Research does not automatically generate knowledge - rethinking product and process

By Catherine Fisher

The overwhelming conclusion I have drawn from my involvement as a contributor to the IDS Bulletin, New Roles for Communication in Development?, is the need for research communication to place a greater focus on process rather than product. I draw this insight from both the process (what I've learned from being involved) and the product itself (what I've learned from some of the great articles in the Bulletin). I'm not arguing that there is no role for product, far from it, but that there is value in re-examining the relationship between the two.

Product v Process? Baking Czech bread
Image credit: Chmee2 (Own work) CC-BY 3.0

Overall I would argue that articles in the Bulletin collectively encouraged us to challenge our assumptions around three related areas:

• What is research? 
• What is knowledge?
• How does change happen?  

And to explore  links between these areas: how does research generate knowledge, whose knowledge produces the research, how does research or even knowledge lead to change?  All speak to a greater focus on process. I explore a couple of examples below.

Research changes those involved 

Patta Scott Villier's article, This research does not influence policy, explores how participation in a research process brought change for both the researchers and the ‘researchees’,  even if the resulting output didn't shape high level policy processes  So the process generated more change than the product. But maybe the product (or its story as described in Patta's article) will inspire someone somewhere else to undergo a similar process, that might itself generate change...

Research doesn’t necessarily build knowledge

In our article, Kirsty, Louise and I explore the factors that shape whether actors engage with research and are willing and able to draw on research.  In doing so, we challenge ideas that research, even if it is shared at the right time and packaged in the right way, will somehow automatically generate “knowledge” in target audiences.  Research does not automatically generate knowledge. This happens through a process of sense-making in which the “knower” is an active participant not a passive recipient, who may deliberately or inadvertently choose to reject the intended message of the product. As Penelope Beynon’s et al's article illustrated, even carefully constructed outputs such as a policy brief, often seen as the silver bullet of research communication, can lead to the creation of different knowledge than was intended.

Growing importance of “process architects”

One implication of this greater focus on process is a greater role for intermediaries, knowledge brokers or innovation brokers within the broad spectrum of research communication. The focus on process over product helps us to see a greater role for these actors which is not just about turning research outputs into attractive products but about seeing research  and research-based products as part of the processes of knowledge creation and change, and supporting the design of these processes. These processes will often draw on products from multiple sources, current or historical, local or from far away. Often these processes themselves produce products which then go on to inform other processes. This speaks to some of the points I make in the Introduction, Is Development Research Communication Coming of Age? (PDF).  

The value and opportunity for exposing the process behind the product

One final reflection is that I doubt anyone will learn as much from the papers that appear in the journal  as I have learned from writing them. The process of writing a paper with a colleague forced us both to reexamine our assumptions and beliefs. An early version of our paper included a box outlining the differences in position about research and knowledge, and how it contributes to change that attempted to share some of the discussions we had  Yet journal articles require a compelling single narrative, an authoritative position and do not allow for divergence and discussion. They also have a strict word count. So this was dropped. While journal articles play an important role in communicating ideas and indeed validating the rigour of those ideas, particularly those from more formal research processes, perhaps there is also space – particularly in the social sciences for us to be more open in our workings, less authoritative in our positions. New media such as blogs like this, or even tweets enable “work in progress/emerging thinking” type products to be shared more widely that can  trigger thinking and knowledge generation processes for others. 

I hope that both this blog series and the Bulletin itself has prompted you to think and re-examine your ideas as it has done for me.

Catherine Fisher contributed to the Introduction in the IDS Bulletin, entitled Is Development Research Coming of Age? (PDF) and to the Bulletin article entitled Stimulating Demand for Research Evidence: what role for capacity building?. Catherine is International Capacity Building Co-ordinator at Amnesty International. She was formally Capacity Support Coordinator with the Impact and Learning Team at IDS.   

More blogs on the IDS Bulletin New Roles for Communication in Development?

• Evolution of new roles: from researcher, to activist, to knowledge intermediary (By Simon Batchelor)
• How researchers can learn to stop worrying and love communicators (By Nicholas Benequista)
• Trying to get research into use? Start by making users an integral part of the research design process (By Abby Mulhall)
• Supply and demand in evidence-informed policy - in pictures! (By Kirsty Newman)
• Three things that stop development organisations being agents of change (By Liz Carlile)
• Why researchers should consider a new model for engagement (By Ajoy Datta)
• An interview with Blane Harvey, co-editor of New Roles for Communication in Development?
• Redefining the researcher, and the research (By Zachary Patterson)
• Challenges in communicating co-constructed knowledge to influence policy (By Fran Seballos)
• How are the roles of researchers and research communicators changing? (By Tessa Lewin)

Evolution of new roles: from researcher, to activist, to knowledge intermediary

By Simon Batchelor

The IDS Bulletin that this series of blogs is associated with was titled New Roles for Communication in Development?  My own article, Changing the Financial Landscape of Africa... was about a ten year journey for a small group of researchers and certainly over the decade (gosh that’s a long time!) our own roles evolved and changed.  

Researcher: In the article I start with some basic research. Commissioned through a competitive bid, we started researching how people were using mobile phones in Africa (and Asia).  This led to an insight about exchanging airtime, where people in three different countries (even in 2001) had worked out that they could do a sort of money transfer using mobile phones and airtime.

Lobbyist/Activist: It is at this point that our role begins to change. We started to move beyond our research responsibilities (as stated on the research proposal) and evolve into ‘research communicators’, even becoming advocates and lobbyists.  The article goes on to describe how conversations occurred with donors, private sector, senior government, central banks – presenting the idea of mobile phone money transfer (and the evidence that there was demand within Africa) and trying to get interest and buy-in to the idea. 

Knowledge Intermediary:  Moving on a year or so our role evolves further.  From lobbying about an idea, we begin to get involved with the details of implementing that idea - legislation, policy environments and private sector development.  Alongside this detail work, we also get involved with creating discussion space for the emerging players.  We start brokering meetings between people, undertaking social network analysis to identify who should talk to who, creating international conferences to create the space for donors, private sector, and legislators to come together, and for bankers and telecommunication to find each other in a new digital converging space.  We were (as Blane et al say in their introduction to the Bulletin (PDF)) “seeking to strengthen the linkages and flows of information between disciplines, areas of practice or sources of knowledge”  So from being a researcher investigating how communities use mobile phones we seem to have evolved into knowledge intermediaries.  

From researcher, to research communicator, to lobbyist, to knowledge intermediary.   Interesting. 

Recently I was working on Theories of Change and Impact pathways for the CGIAR centres, and their proposals really raised the question – where do the responsibilities of a researcher finish?  Do researchers have a duty to change the world, or do they produce their findings and leave it at that?  Do they have a duty to communicate their findings to those who might use them?  Do they have a duty to follow through and make sure that their research does more than sit on a shelf?  What are the boundaries of their and our responsibility?

As a final thought can I just add that the paper is constructed around a framework for policy influence first proposed by David Steven (PDF).  Steven talks about changing the framing of a discourse and creating spaces for dialogue within his five point frame.  I wrote the ‘story’ of the article as a narrative before finding the Steven framework; and then when I found it I realised that the story fits his framing.  I don’t know if all research to policy activism can be mapped onto his frame but I was amazed at how closely he captured our journey – I recommend it.


Simon Batchelor wrote the Bulletin article Changing the Financial Landscape of Africa: An Unusual Story of Evidence-informed Innovation, Intentional Policy Influence and Private Sector Engagement.
Simon is managing director at Gamos Ltd, and formally interim manager of the Impact and Learning Team at IDS. 

 
More blogs on the IDS Bulletin New Roles for Communication in Development?

• How researchers can learn to stop worrying and love communicators (By Nicholas Benequista)
• Trying to get research into use? Start by making users an integral part of the research design process (By Abby Mulhall)
• Supply and demand in evidence-informed policy - in pictures! (By Kirsty Newman)
• Three things that stop development organisations being agents of change (By Liz Carlile)
• Why researchers should consider a new model for engagement (By Ajoy Datta)
• An interview with Blane Harvey, co-editor of New Roles for Communication in Development?
• Redefining the researcher, and the research (By Zachary Patterson)
• Challenges in communicating co-constructed knowledge to influence policy (By Fran Seballos)
• How are the roles of researchers and research communicators changing? (By Tessa Lewin)

How researchers can learn to stop worrying and love communicators

By Nicholas Benequista

Let’s be frank. Researchers don’t really like us research communicators.

They have good reason not to like us, but this doesn’t necessarily need to be so. To explain why, however, I have to go back thirty years or so.

The Development Research Centre on Citizenship, Participation and Accountability (www.drc-citizenship.org) taught us a lot about how researchers connect with political and social actors in distinct ways. Understanding of how individual researchers and their institutions view themselves as agents of change is a good way to begin thinking about a communication strategy, rather than the other way around.  Image credit: N.Benequista

When it was finally acknowledged that policy-makers mostly ignore research, Nathan Caplan postulated the "two communities theory" to explain this shocking phenomenon. He said that there was a "culture gap" between researchers and policy-makers demarcated by their different values, language, professional practices and institutional contexts.

In a sense, those of us who work in research communication are supposed to be the bridges between these two communities, but if you look critically at the literature on research communication, you’ll see that we’ve recreated that culture gap.

On the one side, various academic disciplines - including communication studies, political science and cultural theory - continue to examine research communication for insights into the relationship between knowledge and social change.

This body of mid-range theory, however, has often ignored the politics of the research field, including the increasing pressure imposed on "applied" areas such as development studies or health to deliver "influence." Consequently, these theories are blind to how research communication transforms researchers personally and to how the mainstreaming of research communication is actually a political force. Researchers resent political communicators because we represent an effort change the way they work, and we have to be willing to accept that we have sometimes (though not always) done so for the worse.

On the other side of this divide are the practice-minded guidebooks and handbooks for research communication. These certainly draw on the theory. Some guidebook authors are simultaneously publishing on the topic of research communication in peer-reviewed journals. The majority of them, however, are penned by the agencies - such as the UK Department for International Development or US Agency for International Development - who are mandating that the research they fund shows results. They tend to be bureaucratic and procedural, with long checklists and only the most cursory mention of theory. In my experience, when these “lessons” are delivered to researchers in this format, they are often met with either complete disinterest or with bristling contempt.

And here’s the irony. They are meant to close the culture gap, but they recreate it, because they make no attempt to understand researchers.

The biggest practical challenge of good research communication is a political challenge
Policy network maps and policy briefs are indeed useful skills that need to be learned, but the biggest practical challenge of good research communication is a political challenge of connecting researchers with other actors. The "two communities" theory was later dismissed as it was revealed that many researchers actually inhabit both communities quite well. Indeed, we (Joanna Wheeler and I) would add that researchers occupy an array of different communities - sometimes including high-level policy-makers and other times including local activist groups. But there are, nonetheless, still serious challenges to getting researchers to embrace research communication. This first challenge owes to the very diversity just mentioned. Many are not connected to high-level policy-makers, though perhaps highly influential in other ways: as teachers or as intellectual advisors for social movements. The current best practice for research communication almost completely ignores this diversity.

The second challenge is perhaps even more serious and brings us to the main point, which is the need to approach research communication as part and parcel of the research process. You might call it praxis, or you might just call it integration. Researchers may already be deeply involved in actions to promote political or social change on their own accord, but research communication asks them to do so within the constraints of their professional field. Deep social theory is necessary for understanding the important theoretical questions about knowledge and change raised by research communication, but the practice of research communication poses new professional demands on researchers that must be reconciled with methodological practices and the politics of research.

In other words, research communication isn’t really different from research. This is especially important in development, since it is not the 'ivory tower' that other academic disciplines would aspire to be. This is partly what we’re trying to work towards in our contribution to the IDS Bulletin on New Roles for Research Communication?

So researchers, please stop worrying about the oversimplifications of the policy briefing, and let’s instead think about how we can finally bridge that culture gap.

Nicholas Benequista co-authored the IDS Bulletin article Cartographers, Conciliators and Catalysts: Understanding the Communicative Roles of Researchers with Joanna Wheeler. He is currently carrying out an action research project in Kenya to test the possibilities offered by new communication technologies for journalists. You can follow Nicholas on Twitter at @benequista.

More blogs on the IDS Bulletin New Roles for Communication in Development?

• Trying to get research into use? Start by making users an integral part of the research design process (By Abby Mulhall)
• Supply and demand in evidence-informed policy - in pictures! (By Kirsty Newman)
• Three things that stop development organisations being agents of change (By Liz Carlile)
• Why researchers should consider a new model for engagement (By Ajoy Datta)
• An interview with Blane Harvey, co-editor of New Roles for Communication in Development?
• Redefining the researcher, and the research (By Zachary Patterson)
• Challenges in communicating co-constructed knowledge to influence policy (By Fran Seballos)
• How are the roles of researchers and research communicators changing? (By Tessa Lewin)

Forget asking if policymakers understand evidence – do we understand policy?

By Emilie Wilson

I am setting aside my role as editor for this blog for a minute to share some reflections on a recent workshop I attended. It was called Beyond Communication: exploring approaches to research uptake; and was organised by the UK Collaborative on Development Sciences (UKCDS) and UK Department for International Development (DFID).

Is “Research uptake” more jargon, or a different way of understanding communication? 

The first time I came across the expression “research uptake” was in 2010, when the new UK coalition government was voted in, and a marketing and communication “freeze” across all UK departments was implemented.

I was new to my Communications job at the time, and would have been inclined to agree with my colleague Jeff Knezovich’s observation that “The term was specifically designed to obfuscate its purpose—ironic, given that the whole point was to help clarify and communicate research findings. This obfuscation was a result of a change of government in the UK, and a “communications” witch-hunt from the Conservative-led coalition, branding such activities as “wasteful Labour spending” in a time of austerity”

However, a quick Google search reveals that “research uptake” was around pre-2010 and a number of different sectors have used the term when grappling with the issue ‘how on earth do you get people to do something with the research?’.  In health, for example in this paper “Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science"; or in political economy, for example The Politics of Trade (these examples were thrown up by the ‘quick Google search’ so I’m not holding them up as exemplary merely indicative).

At the workshop, Kirsty Newman – Research Uptake Manager at DFID (and contributor to this blog) – described Research Uptake as “allowing us to take a holistic view” of this issue – here is a reproduction of the sketch I drew in my note-book to capture her description:

The assumption being that previously communicating research findings had focussed on ‘dissemination’ or ‘diffusion’ – i.e. supply; and that research uptake was shifting the focus to ‘demand’ – either ‘what do they want’ or ‘can we persuade them to want what we have’ Or more eloquently described by Jeff as “stimulating an enabling environment among end users of research to commission and find appropriate information to support their own policy processes”. Most people who work in marketing and communications will already be pretty familiar with the focus on stimulating demand, and I’m sure would be only too pleased share their wealth of research on this area!

Evidence-based policy – one way to reduce poverty 

It seems we are still living in an Age of Reason (at least in the northern hemisphere), believing strongly that research validated by expert peers and based on tried and tested methodological approaches has something to offer to people in power. The offer is that research can demonstrate what works or doesn’t work, and it is not supposed to be tainted by money, politics or tribal affiliation – it’s a global public good, a credible source of authority. The assumption is that decisions based on this objective knowledge source will be of the greatest possible benefit, and not privilege a minority or prop up a flawed system.

So, should the ‘demand’ box in my diagram be renamed “creating demand” – i.e. the process is still supply-driven. Evidence (derived from research) – like vitamins, for example – is good for you (and your policymaking) and we need to develop increasingly sophisticated ways for people to engage with it. But do we know much about the demand other than through dialogue which merely engages our ‘end users’ with our research products and processes? Evidence literacy has been used as a term to describe the ability of policymakers and practitioners to understand and apply research evidence, but have we stopped to ask ourselves about our own literacy?

The Kirsty Newman policymaking quiz (PowerPoint)

Kirsty Newman did just that during the workshop, when she put the shoe on the other foot, and asked those present at the workshop some simple questions mostly about the UK policy environment. Such as – what are the three principle functions of the UK parliament? How would you define a “civil servant”? (this was a multiple choice one). I’m ashamed to say that, even with a degree which includes a minor in politics, I only got 4/7 questions correct. And most people present scored less! 

It may have been that people in the room would have been more familiar with other policy environments (and not specifically the UK's), but Kirsty's quiz did well to make the point: do you know the language of the people you are trying to communicate with?

Which is why I think we should consider sticking with the term “communication”, describing a two-way process (from the Latin verb to share) rather than “uptake”, implying a one-way process.

Emilie Wilson is Communications Officer at the Institute of Development Studies, and editor of the Impact and Learning blog. 

Trying to get research into use? Start by making users an integral part of the research design process

By Abby Mulhall

I was recently chatting to Barbara Adolph, a Principal Researcher at IIED, about 'how to get research into use' and we touched on the way that research is commissioned and designed. This led us to talk about our days at the International Centre for development-oriented Research in Agriculture (ICRA), a wonderful organisation that has at its heart the concept of soft systems thinking to design and deliver client-oriented research and development programmes.

What's compelling about ICRA's approach is that it promotes the integration of stakeholder concerns, knowledge, action and learning around a theme of mutual interest. It is about defining the problem with the research users, whoever they were - not just farmers, but researchers, entrepreneurs, journalists, environmentalists - whoever is part of the 'innovation system' that affects research uptake and use.

Defining the research agenda: power and participation

A common thread running through many discussions and debates about research communication (or uptake or use or the many other seemingly interchangeable terms used to define the process of making research available, accessible, useful and useable) is power and participation. While the communication part of research use is essential, there are many other actors, processes and systems that affect the ultimate use of research. Not to mention its availability and accessibility.

Power relations and participation in the research itself, or more importantly in the initial definition of the problem and research design cannot be ignored. There are many examples of research programmes testing more innovative ways of communicating research and some that move further, by using communication as the process for achieving social change (see Nicholas Benequista's and Joanna Wheeler's article in this Bulletin; Jethro Petitt et al in Development in Practice or Sally Theobold et al in Health Research Policy and Systems).

But I think there is a need to re-think how we design development research:

• Who is it really for? 
• Who should determine the agenda? 
• Who should quality assure the results (where the intended users are quite often the poor)?

Liz Carlile's blog in this series touched on a similar point - that we supply knowledge and information on our terms in response to a global conversation rather than local demand. Too often the design of the research programme is done in isolation of the different intended users of the research. They tend to be brought in once the funder has approved the design. Too often researchers have to respond to the policies and processes set by the funder, with little or no time to really reflect, learn and engage in the process of defining the research problem and identifying a solution.

Of course there are exceptions and one of the best examples I know about is CNRS, a Bangladeshi NGO that has worked extensively on natural resources management. Its strength is in inclusive participation and giving voice to a wide range of stakeholders who also have a say in the research agenda. The concept of Citizen Jury, an approach being used to democratise agricultural research, is another excellent example. 

Towards innovation systems thinking

DFID's Research into Use programme is an example of why we need to plan for uptake. Initially designed to get promising technologies from DFID's renewable natural resources programme into use, its learning in developing processes for doing this is extremely valuable.The process is partly about 'knowledge brokering', which is currently very topical and increasingly valued as an essential part of the system (see Catherine Fisher's Spectrum of Intermediary and Brokering functions in the Bulletin's introduction (PDF).

RIU goes a step further and tests an innovation systems approach (see Putting Research into Use: Lessons from contested visions of innovations) - recognising a need for a broader range of brokering tasks to support coordinated action in networks that are connected to innovation, policy and development processes.

In the Bulletin, Klerkx et al write 'research uptake is important, and knowledge brokering is an essential function, but should be accompanied by or integrated within the function of innovation brokering, which more broadly focuses on rearranging all technical, social and institutional relationships needed for innovation and change.Such a broad focus can contribute to creating an enabling environment for effective policy formulation and implementation, development and innovation'. I am a fan of the innovations systems approach to research uptake - though some would question how it addresses power and participation.

On this very topical issue of research communication, uptake and brokering - a new call for abstracts has just been launched - Driving Research Uptake through research brokering.

Abby Mulhall is Research Uptake Manager at the UK Department for International Development (DFID). We are grateful for her contribution to the Bulletin as a peer reviewer. This blog represents the author's views - not those of DFID.

More blogs on the IDS Bulletin New Roles for Communication in Development? 

Supply and demand in evidence-informed policy - in pictures! (By Kirsty Newman)

Three things that stop development organisations being agents of change (By Liz Carlile)

Why researchers should consider a new model for engagement (By Ajoy Datta)

An interview with Blane Harvey, co-editor of New Roles for Communication in Development?

Redefining the researcher, and the research (By Zachary Patterson)

Challenges in communicating co-constructed knowledge to influence policy (By Fran Seballos)

How are the roles of researchers and research communicators changing? (By Tessa Lewin)

Supply and demand in evidence-informed policy – in pictures!

By Kirsty Newman

I have talked before about supply and demand in evidence-informed policy but I decided to revisit the topic with some sophisticated visual aids. I am aware that using the using the model of supply/demand has been criticised as over-simplifying the topic – but I still think it is a useful way to think about the connections between research evidence and policy/practice (plus, to be honest, I am fairly simple!).

You can distinguish between supply and demand by considering ‘what is the starting point?’. If you are starting with the research (whether its a single piece of research or a body of research on a given topic) and considering how it may achieve policy influence, you are on the supply side…

In contrast, those on the demand side, typically start with a decision (or a decision-making process) and consider how research can feed into this decision…

This distinction may seem obvious, but I think it is often missed. What this means in practice is an explosion of approaches to evidence-informed policy/practice which attempt to push more and more evidence out there in expectation that more supply will lead to a better world…

One problem with this is that if your supply approaches focus on just one research project – or one side of a debate – they risk going against evidence-informed policy…

Some supply approaches do aim to increase access to a range of research and to synthesise and communicate where the weight of evidence lies. However, even these approaches are destined to fail if there is not a corresponding increase in demand…

I think we should continue to support supply-side activities but I  think we also need to get better at supporting the demand. So what would this look like in practice?

For me the two components of demand are the motivation (whether intrinsic or extrinsic) and the capacity (i.e. the knowledge, skills, attitudes, structures, systems etc) to use research. In other words, you need to want to use research and you need to be able to do so.

Motivation can be improved by enhancing the organisational  culture of evidence use – but also by putting systems in place which mandate and/or reward evidence use…

Achieving this in practice needs the support of senior decision makers within a policy making institution. So for example the UK Department for International Development has transformed the incentives to use research evidence since Prof Chris Whitty came in as the Chief Scientific Advisor and Head of Research.

But incentives on their own are not enough. There also needs to be capacity and it needs to exist at multiple levels; at an organisational level, there needs to be structural capacity such as adequate internet bandwidth, access to relevant academic journals etc etc. At an individual level, those involved in the policy making process need to be ‘evidence-literate’ – i.e. they need to know whaat research evidence is, where they can find it, how they can appraise it, how to draw lessons from evidence for policy decisions etc etc…

Achieving this may require a new recruitment strategy – selecting people for employment who already have a good understanding of research evidence. But continuing professional development courses can also be used to ‘upskill’ existing staff.

Anyway, the above is basically a pictural summary of this paper in the IDS bulletin so if you would like to read about the same topic in more academic terms (and without the pictures!) please do check it out. Its not open access I’m afraid so if you want a copy please tweet me @kirstyevidence or leave a comment below.

Hope you liked the pictures!

Kirsty Newman co-authored the Bulletin article entitled "Stimulating Demand for Research Evidence: what role for capacity-building?" Many thanks to Kirsty for allowing us to republish this blog, which was originally published on her own blog, KirstyEvidence. You can also follow Kirsty on Twitter. 

More blogs on the IDS Bulletin New Roles for Communication in Development? 

Three things that stop development organisations being agents of change (By Liz Carlile)

Why researchers should consider a new model for engagement (By Ajoy Datta)

An interview with Blane Harvey, co-editor of New Roles for Communication in Development?

Redefining the researcher, and the research (By Zachary Patterson)

Challenges in communicating co-constructed knowledge to influence policy (By Fran Seballos)

How are the roles of researchers and research communicators changing? (By Tessa Lewin)

 

Comparing research and oranges II: do communities want oranges or flowers?

By Simon Batchelor

In her blog, Comparing research and oranges: what can we learn from value chain analysis?, my colleague Elise Wach asks whether “producing research first and then deciding how to communicate it afterwards the same as growing an orange and then deciding how and where it will be sold?” She went on to speculate whether value chain analysis can add something to our own analyses of how to strengthen the knowledge value chain.

Her piece reminded me of a video we used at a team retreats, entitled Whose Reality Counts. Produced by Praxis, based in India, it also caused us to wonder about the comparison with research production, and the processes of setting a research agenda.


In case you don’t have the time to watch the 7 minute video (but please do – it's so well done!) here's a quick synopsis:

A senior office-based person sits and has a bright idea: giving flowers to a poor community. The idea is passed down the decision-making chain to farmers in the poor community, who, initially pleased, begin to plant flowers.

However, still sat in his office, the official continues to pass flowers down the line, and we see the farmer becoming frustrated with too many flowers and not enough diversity of food, which is what the community really wants. The community cannot make their voices heard, until the official goes to the community himself expecting to see grateful villagers and a thousand flowers. That’s not what he finds on his arrival, and it's only after listening the villagers and their needs that he gains an understanding what they really need - their reality as described by them.

For me, the parallels are obvious. Research conducted in isolation from the realities in the field may produce insights, and these initial insights may even be appreciated by the community. However, communities have priorities and there needs to be a feedback loop to find out what those priorities are and whether our research needs to be redirected.

As Elise says in her blog “When is audience research necessary, and when does the ‘if we build it, they will come’ assumption apply? Where is the line between research communication and advocacy? How can we create demand and to what extent should we do so?”

So, whose responsibility is it to set the research agenda?

In a recent review of plans from leading research centres, we had to ask ‘where are the boundaries for a researcher’. If the research centres are intending to change the world in some way (their stated intention) then there needs to be engagement with the outside world during the research.

We ended up noting two type of engagement:

• ‘A need to engage with a representative sample of the end users to ensure that new hybrids or practices fit the ‘real world’ farming systems’. 
• And ‘there are the actors at the boundary of the research who might take the research forward. At some point, research that has led to successful product development will need to go to scale’.

Isolated research may change the world slightly, but may also rapidly become too many flowers when the community needs food. However you frame it – as Value chains with Customer feedback and monitoring market demand, or as participatory development with consultation and ‘mainstreaming the voices of the poor’, research that changes the world is going to require tight feedback loops and a view that is much wider than an agenda set by sitting in an office.