:::: MENU ::::

Putting the tea into theory-driven research

David Blane, Methods

Putting the tea into theory-driven research

Image by Mark Mags. © 2016. © CC0 Public Domain via Pixabay.

Image by Mark Mags. © 2016. © CC0 Public Domain via Pixabay.

By David Blane

As early career health researchers, we IHAWKES have some familiarity with theory, whether it’s a theory of behaviour change from psychology or a grand theory like Marxism or feminism from sociology. I would suggest, however, that when it comes to applying theory to our research, many novice researchers (myself included) are a little less confident. What theory should we use? How do we apply it? Can we generate our own theory?

In health systems research, a variety of theory-driven approaches have been developed in recent years in response to frustration with the perceived limitations of (quasi-)experimental research and evaluation designs, which are great at assessing the effectiveness of certain types of interventions (answering the “what works?” question), but often fail to provide meaningful answers (to the “why?” and “in what circumstances?” questions) when applied to complex systems and interventions. Examples of these theory-driven approaches include theory-based evaluation, theories of change, and realist evaluation and synthesis.

So, should we be using these approaches more often? Or at least learning about them during our research training? The widely cited MRC Framework for Developing and evaluating complex interventions provides broad guidance on the use of theory, but has been criticised for not providing any guidance on how to incorporate theory into the design and evaluation of complex interventions, or examining how the intervention interacts with context.

In my PhD I am using a realist approach to literature synthesis (on interventions aimed at primary care practitioners to improve the identification and referral of patients with obesity) as opposed to evaluation. Realist synthesis is one of the theory-driven approaches that are being increasingly used in the assessment of complex evidence for policy implementation, programs, services and interventions. Yet its application remains somewhat of a dark art, with considerable variation in the way in which it has been applied to different research areas.

To be fair to the advocates of realist methods, they have produced an excellent website and training materials for realist synthesis, with an introductory section on realism and why theory is important. We are told, for instance, that theory helps to focus a review, to determine which literature is most relevant, and provides a guide for interpretation and explanation. So far, so good.

They then go on to describe four different kinds of theory: 1) the underlying philosophy (e.g. realist theory); 2) methodological theory; 3) program theory; and 4) ‘formal’ or ‘substantive’ theory. Am I losing you yet? Because there’s more… “There are three other uses of the term ‘theory’ that are important in realist synthesis. These are ‘initial rough theory’, ‘refined theory’ and middle range theory.”

You can begin to understand why some have suggested that realist methods should only be used by experienced researchers. Personally I disagree, and have found my journey (thus far) into theory-driven research to be extremely satisfying, if a little bewildering at times.

“So where does the tea come in?” I hear you cry…

If I could give one piece of advice to other PhD students or early career researchers considering using a theory-driven approach to their research it would be this – make some protected time and space, with peers or supervisors, to really think through the “what is actually going on here?” question related to your research area. And do it with tea. Lots of tea.

By spending a good amount of time breaking down the detail of what is going on in any intervention, you can begin to separate out the differences between intervention strategies and resources, possible responses to these by all those involved, what the mechanisms that encourage these responses might be, and how these might differ in different contexts (at the individual, interpersonal, institutional, and environmental levels) to lead to different outcomes.

I have written a separate blog on my experience of a recent realist methods summer school in Liverpool, where we had plenty of these key ingredients (especially the tea), and considerable progress was made on a number of different projects. I’d be keen to hear your thoughts on any of the questions raised by either of these blogs. Is your research theory-driven? What has been your experience? Do you know of any good training resources? Comments welcome!

Leave a comment