Huw Davies talked to the LSE Impact of Social Sciences blog on how to treat academic research in order to give it the best foundations when before it enters the policymaking process.
Research Does Not Speak For Itself: research needs to be actively translated in communication; it needs to be set in context, and it needs to be brought to life. By itself, ‘research’ is just inanimate data: in conversation and contextualised it has the power to animate, inform, empower or infuriate. It has the latent capacity to become knowledge, or even evidence.
Which is why I think that when donors ask for indicators of research outcomes they are in fact asking for indicators of the efforts to communicate (share, explain, disseminate, popularise, test, etc) research processes and outputs.
Research Does Not Stand Alone: of course any research must be seen in the context of other research, gradually building up a picture of our social world. Individual studies have far less value than careful synthesis and review. But more than this, research needs to be interpreted in the context of local systems, cultures and resources; and explored with an understanding of political sensitivities, expediencies and implementation challenges.
Which means that policy research programmes should never start with the assumption that nothing can be communicated yet (until we have done the research). Their research is not isolated from other begin done alongside it or done before.
Research Has To Be Integrated: research ways of knowing have to be integrated with other forms of knowing: knowing that comes from a complex and sophisticated conceptual understanding of the world (including ideological preferences), and knowing that comes from deep experience, including tacit ways of knowing or feeling.
We tend to be carried away by the idea that there is one type of evidence: research based evidence. But this is not true. Evidence can be arrived at by different means. Knowledge is even more complex. Whose evidence is as important a question as any.
Using Research is Often Not An Event: use of research is often better seen as a dynamic and iterative process, usually faltering, but occasionally dramatic; most often seen better in retrospect than in prospect. Research-based ideas can slowly seep into policy discourse in a slow and percolative way, gradually changing the sense of what is important or what is possible in policy debates.
Hence the ill-conceived idea of commissioning case studies of specific influencing events, or the stories of change that think tanks tend to use (under pressure from donors). If the process can be easily described within the confines of a brief story of change then it is likely to have been one of those exceptions to the rule -and more often than not the consequence of a consultancy.
It’s Not Just Learning – Unlearning Matters Too: letting go of previously cherished notions, conceptual models or so-called ‘facts about the world’ can be as important as the acquisition of new understandings. But this is far from simple: the new does not necessarily displace the old. Sometimes uncomfortable accommodations or amalgamations are made.
But to do this think tanks need to create sufficient space to test and fail (or make mistakes, at least). The pressure to maximise impact (to ensure that all research is policy relevant and useful -and usable) works against this. Think tanks all over the world offer safe spaces from which to launch new innovative ideas. But this requires a mindset that rewards ideas above all.
Knowledge is Often Co-Produced: rather than seeing research as the preserve of technical experts, new policy-relevant knowledge often comes from collaborative processes that break down the distinction between roles – where technical expertise around data meets other forms of knowing rooted in experience or a sense of the possible. Shared journeys can produce shared understandings.
In a project on trade and poverty in Latin America we commissioned a journalist to study the same issues we were studying but from a different point of view. We never really properly integrated these but it would be a good idea to build coalitions that seek to study the same issues from different disciplines and perspectives. Why not build coalitions between academia, think tanks, NGOs, and the media? A soon to be launched DFID Zambia project will be funding several think tanks to study similar issues and then debate their ideas in public -this has never been seen before!
Knowledge Creation is Deeply Social: the creation of knowledge from various ingredients (including, but by no means limited to, research) is therefore a deeply social and contextual process – happening through interaction and dialogue. It reflects a persuasive process triggered as much in the gut as in the brain.
Dialogue is underrepresented in this sector. We talk about engagement and ‘two-way communications’ a lot but not really about research and influence as a dialogue process. The marketplace metaphor has taken over every aspect of the work including the language with use: demand, supply, etc. We should be talking, as Daniel Ricci says, about a big conversation. Think tanks role there is to improve the terms of the dialogue. Jeffrey Puryear wrote that Chilean think tanks greater contribution was the way they helped politicians to learn how to talk to each other and work together.
Not Products But Process: from all of this it then makes more sense to think of ongoing processes of knowing than the creation and sharing of knowledge products, and so…
And so think tanks contribution is ongoing. A research output or a policy is a step along the way (a not always linear way). Various outputs may contribute to change over time, or prevent it, depending on how they come together. More importantly, though we must not forget that all stakeholders involved in the research and policy process have their own history and agency. They are not static players.
It’s Not All About Decisions But More Often About Framings: because research often has the most profound impacts not when it directly underpins specific decisions (instrumentalist action) but instead when it causes shifts in the language, concepts, conceptual models or frameworks that are used to define the contours of the policy landscape. Research can be at its most powerful when it shakes prior certainties, questions core assumptions or even re-shapes cherished values.
It is not about evidence, it is about arguments!
So, when we focus on research as proving evidence for policy decisions we both overplay its short-term role as technical arbiter and undersell its longer-term transformative power.
I could not agree with him more.
An just so you don’t think Huw Davies is a member of the anti-comms brigade, here is LSE’s bio on him:
Huw Davies is Co-Head of School and Professor of Health Care Policy & Management at The School of Management, the University of St Andrews, and he was formerly Director of Knowledge Mobilisation for the UK NIHR ‘Service Delivery and Organisation’ national R&D Programme (2008-10). His research interests are in service delivery, encompassing: evidence-informed policy and practice; performance measurement and management; accountability, governance and trust. Huw has published widely in each of these areas, including the highly acclaimed Using Evidence: How Research Can Inform Public Services (Policy Press, 2007).