I've just drafted a book chapter on building explanatory theory from qualitative data and now I'm writing a section about developing theoretical coherence as part of a chapter on bringing qualitative analysis to a conclusion - and this question is bothering me. Are there ways in which qualitative researchers claim to be developing theory that do not have an explanatory component?
Also, the terms causal theory and explanatory theory seem to be used fairly interchangeably. Does anyone see a particular difference between these?
I wonder if the following touches on your question. Seems to me that if we treat the idea of the social construction of reality , seriously, eg medical errors, death (Dick Teresi), suicide, etc., then sociologically speaking, theory can offer causal explanations albeit in very different ways from the ways that the natural sciences talk about causality. The latter presupposes that if X causes Y then X must occur prior to Y but Becker's labeling theory , for example, would argue that those labeling actions and actors construct or cause crime; Max Atkinson 's work on suicide would suggest that coroners and medical examiners construct or cause suicides, Teresi I think shows (perhaps against his own thinking) that transplant surgeons and doctors construct or cause death. In other words, social theorizing shows that our practices cause or construct social reality (social phenomena) but presents causality in a non literal ways where it is the practices of the actors that contruct and constitute our world. In short causal explanations in social science are or should be very different from causal explanations in the physical sciences and that when they are the same we are entangled in scientism not explanation.
" it is the practices of the actors that construct and constitute our world"
This is very important. I am also facing the same problem in my PhD research on the Ethnography of Resistance Poetics...where I feel like I am making up resistance where there is none! In other words, in the social construction of reality, who are the partakers, the "actors?"? what is the role of the researcher? I hope this relates to the question of Positionality added to the confusion about "causal theory" and "explanatory theory". And which of the two theories best suit to this kind of activist research where, hypothetically, Creative Resistance (a non-violent action of the marginalized social group against Domination or social injustice) involves folklore and resistance culture--the source where the data come from
I hope this helps rather than adds to any confusion, Asafa, but it strikes me that what I want to do as sociologist is make sense of the sense making of those I am observing. For me that means constructing explanations that take the accounts of the members or actors very seriously; requires that I learn from those whom I observe and not suppose or imagine that I or my discipline have anything to teach those members; and that my accounts of their actions and their understandings make respectful and sensitive use of their accounts without deforming or dismissing or ignoring the good sense they invoke to explain their own actions and accounts and without imposing alternative accounts onto their actions. Those actors are my teachers and I enter their world as a visitor and as a guest and not as a colonizer.
The challenge for all of us then, is to treat as topic the sense making of others while others treat their own sense making as the resource they call upon to engage in and with the world.
It is true " it is the practices of the actors that construct and constitute our world" but very often research as in outsider/imposed narrator gives meanings to these constituted world. Thus, the world actors is quite different from what researcher makes us believe. Researcher cannot be an aloof or uninvolved observer or a tap recorder. In research, you will find researcher everywhere. Even in natural sciences, it is now increasingly recognized that subject-object distinction is only figment of imagination. The rest is explained very clearly by Bernard.
However, I shall request you not to bring the criterion of coherence in qualitative research as it will be death knell of this promising tradition of research. In natural sciences several powerful theories have been sacrificed on this alter. Let a pluralistic colorful research tradition as it is, do not attempt to color all researches in blue.
Hello Pat. Interesting question. I'm not sure whether or not you'll find my reaction helpful; please treat it as a response from someone who is interested in understanding the world so that I and others can act to change it. That may not be relevant to what you're trying to do.
Is theory necessarily explanatory? I'm not being glib when I say I think that it depends on what you mean by "theory" and "explanatory". I assume that the concept "theory" includes the summary of data from a research study. And it seems to me that some theory abbreviates what is happening so that it's small enough to fit in my brain. It helps me to think about what I might be about to do.
(Most of the research I do is intended to be imbedded in immediate action. I use action research, and I expect it may take some trial and error before I find actions that will give the wished-for outcomes.)
Of course I would be well advised to remember that the "theory" is a simplification.
I work in complex situations: communities and organisations. There, I don't find causal explanations (as usually defined) very useful. It often seems that a more useful starting assumption is that almost everything affects (or can affect) almost everything else.
However, my aim is to bring about some change, some improvement, in the situation. The sort of theory I find most useful is what Chris Argyris called a "theory of action". In situation X, to achieve outcome Y, try action Z. (And don't expect it to work without some trial and error.)
Thanks Bob for bringing the discussion back to the kinds of issues my question was intended to raise. While I recognise very much that my goal as a social scientist is to understand and indeed to represent, as best I can, the various viewpoints of my participants, I also believe that if we are to make a contribution to the world, then it is important to develop some coherent understanding of whatever it is we are investigating based on that, that can be of use both to those who contributed that understanding (because it is built on their sense-making) and to others in other places (I too have an action research background and a social justice perspective).
My simple definition of theory, at least at its most basic (local) level, is that it involves a statement of relationships, which I have taken to assume means that it is a statement about this affecting that. I then take a critical realist viewpoint that says it is not enough to just point to the regularity with which this affects that, it is necessary (and useful) also, to understand the mechanism by (or through) which that effect or relationship occurs, and under what conditions that mechanism will operate - with such 'mechanisms' potentially being mental (as people interpret and respond to their world) as well as physical. So, I'm taking a very broad view of theory. And I'm interpreting 'this affecting that' as being 'causal', and consider that developing an understanding of how this affects that is 'explanatory'. In a community improvement setting, theory will be built on trial and error, but hopefully the development of a logic model based on what has been learned (and theorised) from past trial and error will increasingly inform that action so there is increased (theoretical) understanding and less (practice) 'error'. Little theories have the potential to contribute to and eventually grow into bigger theories.
So, I guess my question turns on the issue of whether one can describe some aspect of social reality in a theoretical/abstracted/conceptual sense (i.e., at least a step or two removed from a straightforward reporting of a participant's viewpoint) without implying some element of 'this affecting that' -- or is that just description? ('Just' is not intended to denigrate the value of good description!) I'm kind-of thinking out loud here! And incidentally, the book (Qualitative data analysis: practical strategies) is now published!
Hello again, Pat
Theory as a statement of relationships makes sense to me. And I've been drawn to critical realism at least partly by the same reasons that you offer. It better supports the pursuit of an understanding of the mechanisms behind the relationships. And that includes mental mechanisms.
Let me persist a bit with a theory-of-action approach. I simplified it in my comments yesterday. More fully, it takes the form:
In situation X, to produce outcome Y, try Z, assuming a, b, c, ...
(I expect you know all this.)
I don't think it's too much a stretch to include the postulated mechanisms in the assumptions a, b, c, ....
The key relationship in a theory of action approach is between actions and outcomes:
action --> outcomes [ or more fully: situation ( actions --> outcomes ) ]
And if that's all I have, that's enough. I can act. But to use your terminology (I think), that's little more than descriptive. "In situation X, when I did Y, Z happened."
I think the assumptions mostly sit between action and outcomes. The actions impacted on some of the mechanisms. They in turn acted to generate the outcomes. So it becomes:
actions --> mechanisms --> outcomes
The mechanisms turn description into explanation. In the language of positivist science:
actions --> variables --> variables --> outcomes
There's an earlier and more elaborate attempt at http://aral.com.au/DLitt/DLitt_P03theory2.pdf
(I've ordered your book. It's had glowing recommendations, and I am now waiting for it to arrive.)