Muddling through meanings: doing discourse analysis in health policy research (part 1)

Policy documents are not a collection of “self-evident truths”. Instead, their content reflects a complex interplay of ideas, ideologies, power relationships and contexts. A deeper analysis of these documents, possible through discourse analysis, is necessary to illuminate the role of these factors in the policy process.

In the first of this two-part blog, Ida Okeyo, Sudha Ramani, and Eleanor Whyle, health policy analysis fellows under the programme of the Alliance for Health Policy and Systems Research to strengthen health policy analysis capacity in low- and middle-income countries, discuss their understanding of discourse analysis concepts for health policy research.


Key approaches to discourse analysis
For doing this deeper analysis, we were, during a health policy analysis (HPA) workshop in December 2017, introduced to a group of methodologies collectively termed discourse analysis (DA). In health policy research, DA offers a way of considering the role of language, rhetoric, narratives and framing, which can imrpove our understanding of policy processes (Fischer 2003; Abell, Gruber et al. 2008).


discourse workshop


Discourse as a concept in academia is often closely associated with Michel Foucault. For Foucault, a discourse is an “individualised group of statements” (Foucault 1972) or “groups of utterances which seem regulated in some way and have an internal coherence that distinguishes them from other discourses” (Mills 2004). Understood in this way, DA allows us to explore a wide range of factors including power, politics, and ideas – all things important to policy processes – through language.

DA was a first-time venture for us; and we were excited to try it out. However, we soon discovered its complexities. We began to comprehend that DA is better understood as a broad category of methodological approaches than as any one particular method, and that it can be difficult to grasp the myriad concepts and methodological tools within DA. Below, we briefly discuss our understanding of three approaches: Critical Discourse Analysis (CDA), Framing Analysis and Rhetorical Analysis.

CDA allows for the extension of the unit of analysis from individual words or sentences to naturally occurring social phenomena such as texts, discourses or conversations (Wodak and Meyer 2009). Therefore, CDA is the study of complex social phenomena and requires a multi-disciplinary approach (Wodak and Meyer 2009). In our understanding, the focus of CDA is less on an in-depth line-by-line analysis of the literature and more on the connection of the text with broader ideas in history, society and politics.

Framing analysis has been used to analyse ideas and how they shift over time in policy processes. A frame can be thought of as a central organising idea; a way to organise and represent knowledge and schemes (Koon, Hawkins et al. 2016). Frames are almost always partial and focus attention on some aspects of a political problem and policy solutions, while downplaying others. Hence, more often than not, different policy frames compete with one another (L'Esperance 2013). As a result, framing analysis can also help us to understand the role of power in policy processes, particularly the power of actors and ideas.

Rhetorical analysis focuses on discourse in argumentation and persuasion. This allows for understanding what kinds of reasoning are at play, which can help to reveal underlying ideological and epistemological assumptions that are often hidden or opaque, uncover the struggle over ideas, and illuminate the significance of these ideas in the policy process (Larsen 2004; Russell, Greenhalgh et al. 2008). In addition, because the effectiveness of certain rhetorical devices depends on their resonance with the political or social context, rhetorical analysis can help to draw connections between the policy process and context (Harrington 2009).

While there are many more DA approaches, we felt these three fitted health policy analysis well since they help researchers to explore issues of power, politics, context and ideas. In our understanding, this is what distinguishes DA from simpler forms of content analysis. In content analysis, we explore what is said, but in the above approaches to DA we explore the conditions that made it possible, appropriate and necessary to say it.

Our venture into the world of DA has been interesting, as well as a bit overwhelming. It helped to have the support and encouragement of peers and mentors to collectively process information and clarify our thinking. Having explored different methodologies and approaches within DA, we then faced our next hurdle: how to choose one of these for our research?

Read the second part of the blog: Muddling through meanings: doing discourse analysis in health policy research (part 2)


  1. Abell, J., H. Gruber, G. Mautner and G. Myers (2008). Qualitative discourse analysis in the social sciences, Palgrave Macmillan.
  2. Fischer, F. (2003). Reframing public policy: Discursive politics and deliberative practices, Oxford University Press.
  3. Foucault, M. (1972). Archeology of Knowledge. London, Routledge.
  4. Harrington, J. (2009). "Visions of utopia: markets, medicine and the National Health Service." Legal Studies 29(3): 376-399.
  5. Koon, A. D., B. Hawkins and S. H. Mayhew (2016). "Framing and the health policy process: a scoping review." Health Policy and Planning 31(6): 801-816.
  6. L'Esperance, A. (2013). Fertilize-this: framing infertility in Quebec, Ontario and England between 1990 and 2010, University of Ottawa (Canada).
  7. Larsen, H. (2004). Discourse analysis in the study of European foreign policy. Rethinking European Union Foreign Policy: 62-80.
  8. Mills, S. (2004). Discourse: the new critical idiom. Oxon, Routledge.
  9. Russell, J., T. Greenhalgh, E. Byrne and J. Mcdonnell (2008). "Recognizing rhetoric in health care policy analysis." Journal of Health Services Research & Policy 13(1): 40-46.
  10. Wodak, R. and M. Meyer (2009). Methods for critical discourse analysis, Sage.

Copyright (c) CHESAI