Guidance note 3: Some principles of policy engagement

An emerging consensus

In current debates and writing about effective policy engagement a strong consensus is emerging around a small number of basic principles of good practice, and these have been endorsed in our own discussions with researchers, research support staff and present and former officials of international organisations. Many of the contributions to the debate in this area have had a specifically UK-oriented frame of reference, but in summarising these in this Note we have tried as far as possible to focus on points which will be equally applicable in an international setting. We have also tried to indicate how these principles are closely interlinked, and are to a large extent aspects of a single theme.

Relationships are key

Key points

• Policy engagement will be most effective in achieving impact if it is based on long-term relationships of trust and understanding between researchers and policy actors.
• Early involvement of policy counterparts, for example in the framing of research questions, can help to ensure the relevance and uptake of research findings.

There has been a growing realisation of the relational and social aspects of policy engagement: the most important asset a researcher can have is a well-established, long-term relationship with users and potential users of the research in policy organisations, based on trust and a mutual understanding of each other’s needs and priorities. From the researcher’s point of view this must be founded on a basic appreciation of how the processes of policy-making work in governmental organisations, as discussed in Guidance Note 2, and of the part played in those processes by the people they are talking to. It should also rest on an understanding of what the organisation’s current policy priorities and foreseen needs for evidence are, so that as far as practicable research can be aligned to produce results that are relevant to the concerns of and useable by policy actors. Maintaining regular contact and dialogue can help the researcher to be aware of shifting perceptions and requirements amongst potential users, to take note of emerging issues, and to ensure that policy actors are prepared for and not taken by surprise by emerging findings (especially if they run counter to received wisdom or current policy directions). As many people have said to us, you have to be in the room, to be part of the ongoing discussion through which policy positions gradually evolve.

Maintaining regular contact and dialogue can help the researcher to be aware of shifting perceptions and requirements amongst potential users, and to ensure that policy actors are prepared for emerging findings.

Managing these relational aspects of policy engagement itself requires interpersonal skills – what are sometimes referred to as “soft skills” – which not everybody possesses to the same degree, but which like all skills can be strengthened with practice and good advice. Trust will also be aided if there is complete clarity as to whether the researcher is acting as the advocate for a particular body of findings or interpretation of the evidence, or as an honest broker presenting a synthesis of the current state of knowledge on a topic (this distinction is explored in Dr Ella Adlen’s blog Badgers, bees, beams, floods, and hormones: being an honest broker to policymakers). Trust, moreover, works in both directions: researchers should be able to have confidence that policy actors will use their contribution appropriately, with due acknowledgement and with respect for the integrity of the evidence.

Where possible, and where the nature of the relationship permits, the alignment of research with policy needs can be optimised by involving policy counterparts in the development of research questions; in some cases collaborative relationships may also be reinforced by involving policy people as co-authors of papers. However, some caution is needed in determining how far to go along the path of aligning research to governments’ perceived needs, and not just because policy priorities can change radically at short notice – either because of a change of government or simply of ministers, or because of changing circumstances. The 2015 Nurse Review points out that while applied research “is often best undertaken in partnership with the potential beneficiaries, promoting the co-production of knowledge”, there are risks to the quality of research from pursuing too goal-directed an approach too soon, before the knowledge base will support it: “while a more directed approach might be needed when the problem to be solved is more urgent, it is usually more effective to identify research objectives in a broadly scoped manner, giving freedom for the individual researcher to propose a specific programme within that wider umbrella, and to pursue that research wherever it may lead”.

The cultivation of a longer-term, relational base between researchers and policy actors can help to obviate the situation described by Professor Mark Reed, who reports a civil servant as saying “I’m fed up of being called up by researchers who want to have an impact on me when I’ve got a job to do.” This was echoed by a World Bank official we spoke to, who urged researchers to engage with the Bank first of all through collaboration on its projects and being willing to discuss officials’ concerns with them, before trying to get a hearing for their own research findings; the attitude in the Bank was likely to be, bluntly, “if you couldn’t be bothered to work with the Bank before, why should I listen to you now?” He also warned that colleagues too frequently encountered researchers who were only interested in collaboration until they had secured material for a paper, then dropped away. A former senior International Labour Organization (ILO) official similarly advised researchers to be prepared to be in the relationship for the long haul, perhaps through a succession of small research contracts and collaborations, if they wanted to have real impact. Finally, the establishment of a relationship of trust will help to avoid the risk identified by a former OECD official, that importunate researchers trying to draw attention to their work will be confused with lobbyists for special interests, of whom national and international civil servants are primed to be very cautious.

Relationships pose ethical challenges

Key points

• The weight placed on building relationships and trust as the route to effective policy engagement may make it harder for early career researchers to establish their voice, without specific institutional measures to support them.
• It is essential that relationships with LMIC collaborators are conducted with integrity and sensitivity.
• Researchers must be aware of the value judgements that may be implicit in the way research is framed and conclusions drawn from it, especially when it involves the interests of marginalised groups.

It is important to recognise that stressing the centrality of relationships to effective policy engagement has certain ethical implications, or at the very least raises questions of values. These begin with the question of who gets to form such relationships – who gets into the inner circle of trusted dialogue with policy actors, and how. Many people have suggested the value of a mentoring approach, either informally on a personal basis or through more formal schemes that might be developed by university support systems. This may enable more junior researchers to gain access to existing contacts as a starting point for building their own networks. However, the willingness of senior figures to provide this kind of support and encouragement does appear to vary markedly, partly as a matter of personal inclination and partly sometimes because introducing junior colleagues into these networks may appear to pose a risk to existing relationships of trust and confidentiality. Universities can also take deliberate steps to ensure that opportunities for engagement which they handle are directed where possible to more junior experts, while some funding streams (such as the GCRF) which emphasise engagement with policy-makers in developing countries also aim to involve the best early career researchers. Ultimately, however, credibility in policy circles is based in large part on having a foundation of reputable research (in combination with other factors such as accessibility and a range of appropriate soft skills), so a solid academic reputation will usually be a necessary precondition of policy engagement and impact.

Many of those we spoke to commented on the power of institutional brands in facilitating access to and engagement with policy officials and decision-makers; as one interlocutor said, “if you’re Oxford or the LSE your calls are taken.” These brands of course have considerable reach internationally as well as in the UK. The title “Professor” also carries weight outside academia. Researchers who are in the fortunate position of being able to take advantage of these assets will of course utilise them, but should be aware of the extent to which they may crowd out the influence of less privileged voices in policy debates.

Without some awareness of these issues, and attempts to mitigate them, there is a risk that in any particular policy domain engagement with research based principally on personal networks of relationships becomes or remains the purview of a closed inner circle, to which new or discordant voices find it difficult to gain access, despite the attempts of governments’ formal scientific advisory mechanisms to establish a reasonably broad-based consensus on cross-cutting issues of major concern. This may in turn raise quite profound questions both about whose evidence (and from what discipline) counts in determining specific policy questions, and even about what counts as evidence.

These questions return with even greater force when engaging with policy in low and middle income countries (LMIC). Research in these environments is often facilitated by collaboration with higher education institutions in the countries concerned, which can help ensure access to existing data, and to communities, survey populations, field locations and so on. The most appropriate route to engaging with policy actors in other countries is also often likely to lie through national institutions, whose researchers will have the personal contacts and access that foreigners will otherwise find it hard to achieve. Moreover, collaboration is often a condition of funding. One aim of the GCRF is “to build UK and global development research capacity and capability by forging strong and enduring partnerships between academic communities in the UK and the Global South and by enhancing the research and innovation capacity of both”, and there is a strong presumption that funding proposals will show evidence of such collaboration: “Evidence will be sought that Southern partners – including policy makers and practitioners, as well as researchers – have played a leading role in research design, and planning for implementation and uptake”.

In developing such partnerships it is essential to avoid what has sometimes been referred to as a neo-colonial mindset.

In developing such partnerships it is essential to avoid what has sometimes been referred to as a neo-colonial mindset. One senior figure we spoke to observed that UK academics often behave badly towards their LMIC counterparts, and that more humility is needed; it is important to appreciate the differences in opportunity, possibly training and almost certainly resources which others have to work with. It is also helpful to be aware of the position and status of collaborating institutions; in many countries formerly under colonial rule the establishment of national universities was a significant mark of nationhood following independence, and they retain that prestige. There may be close links between universities and the political sphere, with people from academic backgrounds holding senior positions in government or parliament. In some cases, however, universities or departments within them are seen to be affiliated with opposition politics and may be treated warily by policy actors in government rather than as allies.

It is, however, a question not simply of how relationships with LMIC collaborators are conducted, but also more profoundly of what counts as evidence with regard to policies and programmes that affect marginalised and poor groups – whose evidence is it, and for what purpose has it been acquired? It is more important than ever, when attempting to inform policy in these settings, to appreciate that research and the uses it may be put to are not neutral and value-free, in any discipline, and neither are the apparent solutions to development challenges that they may suggest. Researchers need to see themselves as part of the context, woven into the fabric of power and interests in the countries in which they work and the countries which fund them.

Multiple routes to engagement and impact

Key points

• There are many different tools to use to gain the attention of policy actors – written policy briefs may be effective but are unlikely to be read unless a relationship has already been established.
• Careful consideration needs to be given to which method or combination of methods is likely to be effective, with regard to the nature of the message, the current needs of policy actors, and the culture of the organisation you are dealing with.

There has been an increasing awareness in recent years that the approach to policy engagement which depends on researchers designing and conducting their investigations then attempting to bring their findings to the attention of policy actors – typically by summarising the results and their implications in a policy brief – is unlikely to be the most effective way of achieving impact. This simple linear model is described by Anna Hopkins in her blog “‘Pushing’ research evidence gets us only so far”; although a great deal of effort and funding has been invested over decades in improving research communication and dissemination and making research more accessible to the policy community, it remains a producer-driven “push” model with all the limitations that implies. More recently a greater emphasis has been placed on a relational model of engagement (as discussed above), with more stress on knowledge sharing and co-production and on facilitating “pull” from potential policy users. Alongside this is a concern with building support systems, in universities to assist with policy engagement and in government to improve research uptake. Another way of seeing this is through an adaptation of the old proverb: you can take the research water to the policy horse, the question is how to get the water into the horse.

More recently a greater emphasis has been placed on a relational model of engagement, with more stress on knowledge sharing and co-production and on facilitating “pull” from potential policy users

An important example is provided by the University of Cambridge’s Centre for Science and Policy (CSaP). CSaP’s flagship Policy Fellowships Programme operates on a “policy pull” rather than “research push” basis. Fellows are senior government officials, from the UK and overseas, who are in a position to influence public policy. Some international Fellowships are fully grant funded, mainly through the inclusion of the costs in GCRF grant proposals. The programme introduces each Fellow to 25 academics at Cambridge and sometimes other universities involved in research collaborations to discuss questions which they have submitted as part of their application to join the programme. Because the programme brings the policy actors and their questions to the researchers, rather than the researchers and their answers to the policy actors, what the Fellows gain from the interaction will depend on the academic counterparts’ depth and breadth of knowledge of the topic (and sometimes of methodological approaches) rather than necessarily on the presentation of recent research outcomes. However, the questions that Fellows bring may shape future research directions, and lead to a longer ongoing relationship as a result.

The production of policy briefs is not always an inappropriate approach to policy engagement, but it is most likely to function effectively as a medium of research communication in the context of an already-established relationship with a policy counterpart, or alongside other influencing and engagement activities.

The production of policy briefs is not always an inappropriate approach to policy engagement, but it is most likely to function effectively as a medium of research communication in the context of an already-established relationship with a policy counterpart, or alongside other influencing and engagement activities. A policy brief submitted simply as part of a “cold calling” approach is very unlikely ever to be read either by national civil servants or the officials of international organisations, all of whom have too much to read already. Its fate will be even more uncertain in many LMIC settings; it is unwise to generalise, but in many countries civil servants will be very definite that the culture of their organisation favours oral communication over reading – which extends even to reports they have themselves commissioned. However, where the preference is for oral briefings questions of introductions and access become even more acute. It will also be important to be aware of what the principal working language or languages are in the environment you are targeting; if English is not the most effective means of communication, are you certain enough of being read to justify the effort and expense of getting your policy brief translated (and ensuring the translation does not misrepresent your message)?

DFID’s Research Uptake Guidance, while cautioning against over-reliance on policy briefs or summaries, does also present helpful advice on how to maximise their value.27 It should be said that a short summary has a better chance of being read than a full academic paper in a peer-reviewed journal; while experience shows that copies of particularly relevant papers do circulate amongst colleagues in departments, policy officials are unlikely to have access to pay-walled journals (though research analysts will). Comparable advice is presented in Mark Reed’s Policy Impact Handbook, together with guidance on the use of a range of tools and techniques to communicate research and achieve impact with different groups of stakeholders.

Our discussions with researchers, support staff and officials threw up a wide range of other means of gaining the attention of policy actors. These included:

• Workshops or seminars to present new research; these may be successful in attracting interested policy officials, provided they are short, held in a convenient nearby location, and preferably have a well-known name as speaker or discussant. It is also of course necessary first to have been able to identify the relevant people to invite and been able to contact them.
• Blogs and publication on online platforms with an academic and research orientation such as The Conversation, or the Centre for Economic Policy Research’s VoxEU.
• Social media, maintaining an active professional presence on the usual platforms such as Twitter or Facebook, although it is good advice only to post when you have something useful to say.
• Responding to consultations: providing serious and well-evidenced responses to consultative exercises by government (Green Papers and other consultative documents) or parliament (usually in the form of calls for evidence to assist Select Committee inquiries). Good submissions may have an immediate impact on the development of government’s policy thinking on the topic, or in the case of Select Committees result in an invitation to give oral evidence. While the government is not obliged to implement the Committee recommendations that may result, they can have a very strong influence on subsequent policy development.
• Contact with legislators. Sometimes it may be easier to make contact with and cultivate relationships with individual backbench members with a specific interest in a topic than with government ministers or officials. In the UK parliament there is a vast number of All-Party Parliamentary Groups (APPGs), some with a focus on a particular country and some subject-focused. These have no formal status in parliamentary proceedings, but amongst their members there will be some who have the potential to be genuine and serious champions of an issue or a particular line of research. Similar groupings may exist in other legislatures around the world – such as the bipartisan congressional caucuses or congressional member organizations (CMOs) in the US Congress – but before investing too much time and effort in establishing these contacts it would be advisable for researchers to make sure they know how much influence members may have over government policy in the system concerned. For example, what powers does the legislature, or its committees and backbench members, have to initiate legislation?
• Media appearances. Availability for interviews or comment on current news stories, in the broadcast or print media, can enable researchers to establish visibility and a personal brand, and pave the way for their own original work to be noticed in policy circles. Universities can help to channel media requests to the most appropriate person. However, media work can be time-consuming and (for some) stressful, and leave researchers with little control over how they or their work is presented. Conversely, policy organisations may sometimes approach academic experts they know or have worked with for a quick response to press coverage that affects their interests. One senior researcher we talked to spoke of the need to try to steer government departments, here and internationally, away from rapid responses to press stories and short-term consultancy projects and towards more strategic research that will establish a solid evidence foundation for future policy.
• Conference attendance is an essential way of building networks as well as exchanging knowledge with professional peers, and of being part of the discussion, but not all such events will attract people from policy organizations and lead directly to engagement.

Some of these approaches are geared towards directly disseminating the results of current or recent research, others more towards establishing personal visibility or shifting the terms of debate, the way in which issues are framed in the public understanding, so as to pave the way for more direct engagement with policy processes in future. Different (combinations of) approaches are likely to be effective in different contexts, and some will be more feasible than others in international settings; it is worth spending time considering what the policy actors you hope to influence actually want, both in terms of the material they are likely to be interested, and the format that is most likely to gain their attention. Stakeholder analysis may help to think systematically about who in the policy sphere is likely to be interested in your topic, and who is likely to be able to influence decisions.28It is also worth bearing in mind, in deciding where to place effort, that with any online activity there is a huge signal-to-noise problem. Quite apart from the general debris that strews the world of social media, even among serious, high quality blogs there is such a profusion of constantly-renewed material that it is hard for any message not to get drowned out.

Different approaches are likely to be effective in different contexts, and some will be more feasible than others in international settings; it is worth spending time considering what the policy actors you hope to influence actually want, both in terms of the material they are likely to be interested, and the format that is most likely to gain their attention. 

Finally, the reasons why policy impact is hard to measure in any concrete way have already been established. It is therefore equally hard to measure the absolute or relative effectiveness of different routes to achieving engagement with policy actors. Perhaps the most practical metric is one suggested by one of our interlocutors, and in essence an input measure – “impact is the time they will give you.”

Resources and support

Key points

• Policy engagement, whether establishing contact with potential policy users in advance of a grant proposal, maintaining ongoing dialogue or communicating research results, entails costs of both time and money.
• Achieving impact is likely to require persistence, but handled insensitively this may become counter-productive.

Establishing contacts and networks

The emphasis on cultivating long-term relationships with policy actors, building networks, and maintaining informal contacts and dialogue as well as securing more formal research or consultancy contracts where possible, has practical and resource implications. “Being in the room” has costs, both in time and money, which early career researchers may find it difficult to accommodate. While a senior figure can comfortably refer to spending an afternoon at the Treasury to provide policy suggestions as an example of one of the less formal paths to impact, such engagement may be much more difficult for junior colleagues, given the need not only to juggle the competing demands on time of research, teaching, grant applications, publication and administrative tasks, but also the need to afford train fares and possibly child care (it is another ethical aspect of engagement that the latter is still more likely an issue for women academics). Workshops and other networking or dissemination events also have costs, even if modest – the price of a venue, refreshments, printed material and travel expenses for speakers or discussants – while conferences can entail quite substantial outlay for attendance fees and accommodation and travel costs.

The process of establishing initial contact with key players in governments and parliaments both in the UK and internationally, and indeed in research institutions abroad, may be greatly lubricated if those you wish to build a relationship with are alumni of and favourably disposed towards your own institution. Many influential figures in government circles around the world are graduates of Oxford and other leading British universities. However, the potential use of established alumni networks to identify such contacts raises considerable sensitivities – not least of data protection – and needs further consideration before it can be recommended.

Some engagement costs can and should be built into grant proposals, and this is indeed essential if researchers are to plan from the outset how their work may be relevant to and have impact on policy, including consulting potential users in shaping the proposal.

Some engagement costs can and should be built into grant proposals, and this is indeed essential if researchers are to plan from the outset how their work may be relevant to and have impact on policy, including consulting potential users in shaping the proposal. In the case of the GCRF, proposals are explicitly required to provide evidence of the existence of networks including policy makers or credible plans to grow and sustain these; as part of the GCRF the Scottish Funding Council has provided a Travel and Partnerships fund to enable researchers to develop links with LMIC partners, with a view to facilitating both research proposals and long term collaborations.

Sustaining relationships

The question remains, however, of how communication and engagement activities can be sustained after the end of a grant period – people we have spoken to from both the researcher and policy sides of the dialogue have commented on the importance of persistence and a long-term commitment to engagement. Persistence is often needed to break through initial wariness and resistance to new approaches in bureaucracies, to identify and get access to the right people to be talking to, and gradually to shift the terms of debate: this takes time and may need money as well (particularly for travel). It is good advice, though, to persist but not pester! Given the pressures of time, resources and a clamour of competing information that policy actors very often work under, once a researcher has become a nuisance to be avoided rather than a potential source of help the battle for impact has been largely lost.

Testimonials are an important source of evidence for impact in REF case studies, and will be easier to request in the context of an ongoing working relationship. Nevertheless, the limits of what can be expected at least from UK government departments is well expressed in the updated FCO ARI: “Asked for REF feedback or statements of support for funding bids, our input will in most cases be a simple, factual statement noting the nature of the researcher’s interaction with and contribution to an FCO team (for example the fact that the academic spoke at a seminar for policymakers on a particular topic) and outlining the relevance of the event, endeavour or proposed topic for FCO strategic objectives.” Obtaining useful feedback may be proportionately harder to obtain in international contexts, where the rules of the UK research funding game are less well understood, and may require some investment of effort.

There is a widespread recognition of the need for even quite small amounts of seed funding to help support all of these engagement activities. Some possible sources of support are mentioned in Guidance Note 7.

Managing the career path

Key points

• Policy engagement and impact are of huge importance to academic careers, but priority needs to be given to formal publication of excellent research papers.
• What policy actors are often most interested is the academic’s accumulated experience and expertise as a source of authoritative advice on the issues that concern them, and that credibility is founded in large part on a record of publication.
• REF impact case studies also have to be linked to excellent research.

Balancing policy engagement and other imperatives

Given the level of commitment of time, energy and resources that is often needed to achieve policy impact, researchers need to give careful and deliberate consideration to how much prominence they give to this aspect of their career development.

For many there is a strong intrinsic motivation to engage with policy, in addition to the employment requirements of institutions, and the powerful extrinsic motivations provided by the criteria for some grant funding and the need to demonstrate evidence of impact for the REF. There is, however, a very widely acknowledged tension between the demands of academic publication in reputable peer-reviewed journal and the time that can be taken up by policy engagement work, which is felt especially acutely by early career researchers. Indeed, some of the senior academics we spoke to were anxious to caution junior colleagues against getting drawn too deeply into engagement work too early. “Write the excellent research paper first” appears to be the best advice.

While the design and conduct of research projects that can be expected to have policy impact will benefit from early and frequent dialogue with potential users, and the cultivation of networks of researchers and practitioners, ultimately the authority and credibility of the researcher is founded on the quality of the published work. Once that has been achieved greater effort can be devoted to communication, dissemination and the persistent pursuit of impact – whether through subsequent summaries and policy briefs or other means. Even for the REF, the impacts claimed in case studies have to be attributable to excellent research.

Some possible pitfalls

There are some other pitfalls in the early assiduous pursuit of engagement at the possible cost of serious publication.

The REF procedures provide for the appropriate handling of confidential material in case studies, but that scarcely meets the point that indiscretion in revealing the content of internal discussions will be fatal to the vital relationship of trust.

First, although working with government may provide access to valuable data, the closer a researcher gets to the real nexus of policy discussion and gestation of advice for ministers within departments, the more likely it is that what they learn will be confidential and not disclosable or citable in subsequent publications. The REF procedures provide for the appropriate handling of confidential material in case studies, but that scarcely meets the point that indiscretion in revealing the content of internal discussions will be fatal to the vital relationship of trust. What the researcher learns in these interactions may be extremely interesting and informative from a professional point of view, but not easily translated into publishable insights.

The second concerns the nature of the commodity that is being traded in knowledge exchange transactions. What the researcher may want to “sell” is the discrete piece of new knowledge emerging from their research, which they perceive to have potential value as evidence to inform public policy, and through which they hope to demonstrate impact. What policy actors in governments or international organisations may be more interested in “buying” is the accumulated subject knowledge and experience of the academic expert, his or her tacit knowledge and expertise, as a source of authoritative advice on current problems or issues.

These prospective mismatches in expectation can be negotiated, especially if the researcher is willing and able to be flexible and is prepared to discuss and help policy officials with their challenges on their terms; this then provides the platform from which they can market their own knowledge products. However, the credibility and authority which officials will most value and seek out – and which gives weight to the opinions which they can cite in the advice they put forward – stems ultimately from a record of high-quality publication. This again reinforces the idea that – while it is not advisable to wait until research is completed before thinking about how it might be of use to the policy community – primacy should be given to the demands of academic publication. Journal papers may not be read by policy actors, but they provide the reputational basis for later engagement (possibly mediated by specialist research analysts in departments, who are more likely to read the leading journals and recommend authors to approach).

A final, unserious word on the possible risks of policy engagement: In the children’s story, Chicken Licken fails to get her empirical evidence that the sky is falling to the attention of the appropriate royal decision-maker – and just as well, as her research is methodologically flawed, and her results lack independent corroboration and are not replicable. The ill-considered attempt at engagement not only fails to achieve impact; it is also terminally damaging to her career and that of her associates.

Some questions to consider

• How relevant to any current public policy concerns is your field of research? What links do you already have to people concerned with the development or implementation of policy? What are the policymakers looking for, what are their questions?
• How could you go about establishing new contacts of this kind? Is there anyone who might play a mentoring role and introduce you to their network? Is there anyone you could play a similar role for?
• How much time and effort do you plan to devote to pursuing policy engagement? When would it be appropriate to do this?