How Should Academics Engage in Policymaking to Achieve Impact?

This article reviews the advice from the academic and ‘grey’ literatures to identify a list of dos and don’ts for academics seeking ‘impact’ from their research. From ‘how to do it’ sources, we identify consistent advice on how to engage effectively, largely because it is necessarily vague, safe, and focused primarily on individuals. We then consider the wider policymaking system in which actors make political choices and have unequal access to impact opportunities. We identify the effort it takes to have actual policy impact and how far academics should be expected to go to secure and take credit for it.


Introduction
Academics are under increasing pressure to engage with policymakers, practitioners, publics, and traditional and social media. However, they face major ethical, personal, and practical dilemmas about if, when, and how to engage to influence policy. Furthermore, the positive and negative effects of such engagement, from workplace promotion to personal intimidation and social media abuse, are not experienced equally. In that context, there has been an explosion of activity, on the theme of dos and don'ts for academics, but with no guarantee that 'one best way' to engage for policy impact will ever emerge.
What does the currently available advice add up to? Does it produce consistent messages that can be organised into key general themes for all academics, with the potential to be tailored for political studies researchers in a straightforward way? Or, is the advice based on narrow points of view from specific individuals or disciplines that are not relevant to political studies? Does it help academics secure meaningful 'impact' or merely help them play the game and describe enough impact activity to satisfy their employers and funders?
To help answer such questions, we first draw on systematic reviews of two sources of general advice on impact: (1) peer-reviewed articles by scientists describing their experiences of the 'barriers' between evidence and policy (Oliver et al., 2014a) and (2) the 'grey' literature, in which there is a rich source of reports and blogs by experienced researchers, practitioners and policymakers (Oliver and Cairney, submitted). From these sources, we can identify fairly consistent advice that is relevant to political studies scholars. For example, most accounts emphasise the need for short, concise, and freely available reports in plain language, to counter a tendency towards inaccessible jargon-filled articles behind a paywall. Further, many encourage more face-to-face contact with policymakers and practitioners, to help us understand and tailor our research to our audience, while some advocate the greater use of blogs and a professional social media presence.
While such advice seems sensible, it is not informed routinely by policy studies or political science accounts of the relationship between evidence and policy (Cairney, 2016). Consequently, there is a problematic tendency to produce advice that is too general, on the assumption that advice applicable to one type of scientist is applicable to them all; too 'safe', without exploring the politics of engagement; and, too reliant on a linear idea of impact in which there is a direct relationship between activity and outcome. Social science accounts question the idea that academics can apply such generic advice to have such a direct effect on policy and policymaking (Boswell and Smith, 2017). Rather, game-playing Universities use this understanding to tell an overly heroic story of individual academics (Dunlop, 2018; see also Moran and Browning, 2018). It is important to separate some general, sensible, 'how to' advice regarding activities like clear communication and networking from the more specific and challenging advice -regarding concepts such framing and coalition-forming -that we would associate with political activity and derive from actual studies of evidence-informed policy change (Cairney and Kwiatkowski, 2017). Indeed, it would be ironic if political studies academics restricted themselves to the safe but often low-impact or unreflective strategies recommended by their peers in other disciplines. Therefore, to challenge and help improve the 'how to' advice-found predominantly in the 'grey' literature and periodically in peer reviewed 'barriers' studies-we identify the ways in which academic political science and policy studies raise key issues and frame more fundamental questions. First, the 'how to' literature focuses on individuals or organisations, such as when recommending concise reports and a social media profile. Policy studies focus more on policymaking systems and the difficulties of separating the effect of individual action from systemic effects. Effective actors, in such systems, tend to invest for the long term to, for example, become part of larger coalitions and learn how to frame evidence in relation to the beliefs of their audience. Second, few sources of advice address ethical or political dilemmas regarding, for example, variations in the power and vulnerability of researchers when they engage in politics and policy. Key issues can range from the line we think we can draw between evidence framing and manipulation (Cairney and Oliver, 2017), the balance between tailoring advice and pandering to the ideology of our audience (Cairney, 2018a), and the extent to which Universities can expect academics to engage on social media when they know that some may be listened to less but abused more.
Third, these issues intersect with systemic issues regarding what it really takes to have policy impact. Put most strongly would we expect academics to engage for the length of their career-while often feeling confused, vulnerable and compromised-in the hope of exploiting a 'window of opportunity' for change that may never come? Or, put more generally, to go beyond standard advice is to consider the wider policymaking system in which academics must make political choices and exercise power, raising more profound questions about what it takes to have impact and how far academics should be expected to go to secure it.
We explore these issues in the following sections, asking how each source of advice engages with safe issues versus key dilemmas. First, we synthesise insights from two sources of literature on impact and evidence-informed policymaking: peer-reviewed studies of the 'barriers' between evidence and policy, generally from health and natural sciences; and, lessons on engagement from the grey literature, written by experienced practitioners and researchers of the policy process. Second, we show how studies informed by policy studies and political science change the way we think about impact, from a focus on individual advice to systemic issues and dilemmas. Third, we reflect on the gap between safe advice on how to engage and the more challenging issues that arise when we consider what it would take to secure real, long term policy impact with evidence. Overall, we reject the idea that political scientists can draw on generally applicable 'how to' advice. Furthermore, political and policy studies concepts help us identify the major dilemmas that scholars face when they seek to engage for impact.

Studies of the 'barriers' to academic impact and practical advice on how to respond
In the UK, a key source of context is the relatively hopeful story of academic impact contained in the 'Research Excellence Framework' and the requirement to describe 'pathways to impact' to secure Research Council funding (Boswell and Smith, 2017: 2). It builds largely on 'linear models of the policy process, according to which policy-makers are keen to "utilise" expertise to produce more "effective" policies' (Boswell and Smith, 2017: 1). If so, governments will pursue 'a more "evidence-based approach to policymaking", researchers 'have a responsibility to articulate the impact of their research to nonacademic audiences', 'this impact can be documented and measured', and 'researchers' own efforts to achieve research impact will play a significant role in explaining why some research has impact beyond academia and some does not' (Boswell and Smith, 2017: 2-3). Boswell and Smith (2017: 7) suggest that such 'simplistic supply-side models' may 'offer a reassuring narrative to both policy-makers and researchers' but do not provide the types of 'theoretically informed' analysis that would help us waste less time, play fewer games, and think more sensibly about impact. Rather, they exacerbate two key problems in the literature.
First, a cadre of scholars of evidence/policy draws incomplete conclusions when trying to explain the main 'barriers' to their impact. Oliver et al. (2014a) conducted a systematic review of 145 articles published since 2000 on the 'barriers of and facilitators to the use of evidence by policymakers'. Most focus on health, generally providing insights from the perspective of researchers, and often with a comparison between 'evidence-based policymaking' and 'evidence-based medicine' in mind (Cairney and Oliver, 2017;Oliver and Pearce, 2017). Very few studies draw on theoreticallyinformed analysis of the policy process (Oliver et al., 2014b). Cairney (2016: 90-92) supplements this search in the field of environmental science, finding a similar focus on personal experience or surveys of scientists describing the obstacles they faced. These limitations are reflected in the proposed solutions to key barriers, including the following: 1. Produce better quality evidence on policy problems and solutions. 2. Improve dissemination strategies to increase policymaker access to research: write more concise and less jargon-filled reports, boost resources for dissemination and remove paywall obstacles to accessing research. 3. Develop relationships with policymakers, to address the unpredictability of politics, or the importance of timing, serendipity, and 'windows of opportunity' to act. 4. Engage directly, in academic-practitioner workshops, or use intermediaries such as 'knowledge brokers', to break down communications and cultural barriers associated with the different incentives, rhythms and language of researcher and policymakers. 5. Encourage policymakers to be more science literate, to appreciate the role of evidence and ways to separate high-and low-quality sources (Cairney, 2016: 57-58, 90-92;Oliver et al., 2014a;Topp et al., 2018).
Second, there is continuous anxiety among researchers asked to do the impossible with their research using 'how to' advice found regularly in the 'grey' literature. Oliver and Cairney (2019) searched systematically -in Scopus, Web of Science, Google Scholar, and selected websites and journals -for academic, policy, and grey publications which offer advice to academics or policymakers on how to engage better with each other. This search captures letters, editorials, think-pieces, and blogs, all of which are usually ignored by evidence syntheses. These sources produce a remarkably consistent set of tips over time and across disciplines. We summarise the key themes and individual recommendations from 86 publications (see Oliver and Cairney, 2019 for a full account of method and results): 1. Do high-quality research.
2. Make your research relevant and readable.
7. Be 'entrepreneurial' or find someone who is.
• • Much advice projects an image of a daring, persuasive scientist, comfortable in policy environments and always available when needed. (Sebba, 2011) to 'sell the sizzle' (Farmer, 2010).
• • Become able to 'convince people who think differently that shared action is possible', (Fischoff, 2015) and that real, tangible impacts are deliverable ). • • If not able to act in this way, hire brokers to act on your behalf (Marshall and Cvitanovic, 2017;Quarmby, 2018).
8. Reflect continuously: should you engage, do you want to, and is it working?
• • Academics may be a good fit in the policy arena if they 'want to be in real world', 'enjoy finding solutions to complex problems' (Echt, 2017a;Petes and Meyer, 2018), or are driven 'by a passion greater than simply adding another item to your CV' (Burgess, 2005). • • Keep track of when and how you have had impact, and revise your practices continuously .
It is difficult to conclude that these solutions would boost research impact significantly, largely because they are based on questionable diagnoses and remain unlikely to happen (e.g. boosting science literacy in policymakers), or because they only address one part of a larger problem (e.g. communicating simply). For example, there is minimal focus on the competition to define good evidence. Most policymakers -and many academics -prefer a wide range of sources of information, combining their own experience with information ranging from peer reviewed scientific evidence and the 'grey' literature, to public opinion and feedback from consultation (Cairney and Oliver, 2017;Davidson, 2017;Lomas and Brown, 2009;Nilsson et al., 2008;Nutley et al., 2000Nutley et al., , 2013Weiss, 1979). In that context, the task is not simply to summarise concisely what you think is the best evidence, but also to frame its implications to make it policy relevant and in demand by policymakers (Topp et al., 2018).
Furthermore, there is insufficient focus on the factors that political scientists and policy process scholars would use to understand the role of evidence in policymaking: the ways in which policymakers address 'bounded rationality', and dilemmas created by a complex policymaking environment in which the discrete effect of individual action is often impossible to determine (Andrews, 2017;Cairney, 2017aCairney, , 2018bCairney et al., 2016;Cairney and Kwiatkowski, 2017;Cairney and Oliver, 2017;Cairney and Rummery, 2018;Cairney and Weible, 2017;Cairney and Yamazaki, 2018;Jones and Anderson Crow, 2017;Parkhurst, 2017;Sohn, 2018;Witting, 2017;Zampini, 2018).

How do theory-informed policy studies challenge this advice?
Policy studies provide two profound qualifications to the dos and don'ts literature. First, they seek to capture the effect of 'bounded rationality' which -in comparison to the ideal-type 'comprehensive rationality' -describes the inability of policymakers to separate their values from facts, rank their preferences consistently, make policy in linear 'stages', or analyse the policy and policymaking context comprehensively (Lindblom, 1959;Simon, 1946). Although advances in scientific method and technology might appear to help solve this problem, they do not. The 'radical uncertainty' of policy problems makes them difficult to predict (Tuckett and Nikolic, 2017). More evidence does not help us adjudicate between unclear preferences or simplify the policy process in which they are considered. Policy-relevant science remains value-laden, from the decision to ask a specific research question on a problem in a specific way, to the ways in which we evaluate the success of a solution (Botterill and Hindmoor, 2012;Cairney, in press;Douglas, 2009).
Although it is possible for scientists to ignore wider debates on their own knowledge claims and values, they must contend with the bounded rationality of policymakers. Indeed, there are profound consequences to the ways in which policymakers deal with it. Cairney and Kwiatkowski (2017) describe cognitive shortcuts provocatively as 'rational', to use simple rules (including trust in expertise) to identify good enough sources of information, and 'irrational', to use their beliefs, emotions, habits, and familiarity with issues to identify policy problems and solutions (see Baumgartner, 2017;Haidt, 2001;Jones and Thomas, 2017;Kahneman, 2011;Lewis, 2013;Sloman and Fernbach, 2017). Yet, the key point is the focus on cognitive shortcuts overall, to turn our understanding of the role of policy-relevant research evidence on its head. Ditch the idea of a 'knowledge deficit' to be solved by more scientific evidence in the pursuit of comprehensive rationality (Crow and Jones, 2018). In its place, embrace the image of policymakers seeking 'computationally cheap' (Gigerenzer, 2001) ways to ignore almost all evidence to allow them to make choices decisively.
This focus on cognitive shortcuts helps us understand the ways in which effective actors present information to influence policy: a narrow scientific emphasis on producing more information to reduce scientific uncertainty should be expanded to a wider emphasis on framing research evidence to address ambiguity (Cairney, 2016;Wellstead et al., 2018). Ambiguity generally describes disagreement on how to interpret the world, and specifically the many ways in which we can describe an issue as a policy problem. Actors compete to draw attention to one 'image' of a problem at the expense of all others and, if successful, they limit attention to a small number of feasible solutions (Baumgartner and Jones, 1993;Kingdon, 1984;Majone, 1989;Zahariadis, 2007). Indeed, the competition to resolve ambiguity helps determine the demand for evidence. This is a highly political process to exercise power to determine who describes the world and its most important problems, not a technical process, to research naturally important issues objectively without considering how we define them.
Second, theory-led studies examine the implications of policymaking complexity. The classic ideal-type of policymaking identifies a policy cycle containing a series of welldefined and linear stages (Wegrich and Jann, 2006: 44). In this scenario, we know when and how to present evidence, to help measure the size of a problem (agenda setting), generate evidence-informed solutions (formulation), and use evidence to implement and evaluate solutions before deciding if they should continue. This image is one of the few described in the 'barriers' literature (Oliver et al., 2014a) but is, at best, a story for policymakers to tell about their work, not an accurate description of it (Cairney, 2015;Topp et al., 2018). In contrast, policy theories combine key political science concepts to capture the constituent parts of policymaking environments, summed up as follows (Cairney and Heikkila, 2014: 364-366;John, 2003: 488): 1. Many actors making and influencing choices at many levels of government.
Researchers are competing with many actors to present evidence and secure a policymaker audience, and there are many 'venues', or arenas in which authoritative decisions can take place. 2. A proliferation of 'institutions', or the rules and norms maintained by many policymaking organisations in each venue. Studies of 'new institutionalism' suggest that these rules can be formal and well understood, or informal, unwritten, and difficult to grasp (Ostrom, 2007a(Ostrom, , 2007b. They include the many possible rules of evidence gathering, from who takes the lead to the sources and types of evidence they favour. 3. The pervasiveness of policy networks, or the relationships between policymakers and influencers, many of which develop in 'subsystems' and contain relatively small groups of specialists. 4. A tendency for well-established 'ideas' -as the 'core beliefs' of policymakers or 'paradigms' in which they operate -to dominate discussion (Hall, 1993). They provide context for policymaking, influencing levels of receptivity to new policy solutions proposed to policymakers (Kingdon, 1984). 5. Policy conditions and events that can reinforce stability or prompt policymaker attention to shift. Social or economic 'crises' or 'focusing events' (Birkland, 1997) can prompt lurches of attention from one issue to another.

Seeking impact: When safe advice meets professional dilemmas
These concepts describe a wider context in which to gauge the meaning and applicability of practical advice, in three main ways. First, note the extent to which general 'how to' advice would change with these factors in mind. If there are so many potential authoritative venues, devote considerable energy to finding where the 'action' is (and someone specific to talk to). Even if you find the right venue, you will not know the unwritten rules unless you study them intensely. Some networks are close-knit and difficult to access because bureaucracies have operating procedures that favour some sources of evidence. Research advocates can be privileged insiders in some venues and excluded completely in others. If your evidence challenges an existing paradigm, you need a persuasion strategy good enough to prompt a shift of attention to a policy problem and a willingness to understand that problem in a new way. You can try to find the right time to use evidence to exploit a crisis leading to major policy change, but the opportunities are few and chances of success low. In that context, policy studies recommend investing your time over the long term -to build up alliances, trust in the messenger, knowledge of the system, and to seek 'windows of opportunity' for policy change -but offer no assurances that any of this investment will ever pay off (Cairney, 2016: 124;Cairney and Oliver, 2017;Stoker, 2010;Weible et al., 2012). This advice can also be found in parts of the grey literature, which suggests that not everyone has the motive or skills to be effective in persuasion or network formation. Influencing policy is 'a specialist, time-consuming activity' (Lloyd, 2016) that takes huge cognitive and emotional labour, often requiring community and institutional support . Addressing, but not solving, complex problems with real-world ramifications should excite you, perhaps to the extent that entering public service is the most likely route to impact (Farmer, 2010;Petes and Meyer, 2018).
Second, note how this wider policymaking context -and the weak link between engagement and payoff -informs discussions in the grey literature about common dilemmas:

•
• Should academics advocate for policy positions  and offer policy implications from their research (Goodwin, 2013)? Or, should they be careful not to promote particular methods and policy approaches (Gluckman, 2014;Hutchings and Stenseth, 2016;Prehn, 2018), leaving this political role to specialist scientific advisors (Hutchings and Stenseth, 2016), to maintain academic independence and impartial advice (Alberts et al., 2018;Dodsworth and Cheeseman, 2018;Whitty, 2015) or reduce conflict (De Kerckhove et al., 2015)? The academic literature on policy networks and communities suggests that policymakers and influencers engage so regularly that they adapt to each other's beliefs, and often begin to share an outlook on the policy problem, which blurs the boundaries between formal authority and informal influence (Jordan and Cairney, 2013). Therefore, the dichotomy between engaging to provide advice versus recommendations becomes artificial; successful evidence advocacy requires a level of engagement in networks that blurs the divide between scientist and policymaker (Himmrich, 2016). • • Should academics help 'co-produce' knowledge and policy with others?
Co-production is often hailed as the most useful way to promote research evidence in policy  but, to do so in a meaningful way, researchers must cede control over the research agenda (Flinders et al., 2016;Matthews et al., 2017). There are reputational risks: it can prompt accusations of bias, partisanship, or partiality for one political view over another. Yet, the implicit or explicit framing of these risks is often in relation to (a) an artificial image of the academic as impartial, and/or (b) the idea that academics have other, more straightforward, options to achieve policy impact. Alternatively, if we accept that impact requires a more profound level of engagement, we see that the risks to co-production are no higher than other feasible strategies. • • Should academics engage for instrumental reasons or engage in more sincere engagement, perhaps even to change their outlook and improve their research? Much advice rests on the assumption that academics are engaging primarily to persuade policymakers to privilege and act on their research. A better choice is to engage primarily to listen and learn, then reflect on their research practices, outputs, and most useful contribution (Parry-Davies and Newell, 2014). Indeed, the instrumental academics may be damaging the relationships and goodwill built by the more sincere and invested participants who possess a more enlightened view on the likelihood and nature of their impact (Goodwin, 2013).
Third, policy studies and political science concepts help raise specific issues about the dilemmas associated with impact. For example, 'new institutionalism' studies help us understand the profoundly unequal payoffs to engagement. Broadly speaking, historical institutionalism might suggest that evidence may not influence policy dramatically if it goes against a series of choices made and reinforced over decades (Pierson, 2000), while discursive institutionalism suggests that policymaking is more open to the types of communicative discourse that may suit some social scientists (Schmidt, 2010). More notably, feminist institutionalism suggests that the 'rules of the game' in politics provide unequal access to men and women (Chappell and Waylen, 2013;Kenny, 2007;Lovenduski, 1998;Mackay, 2004). Furthermore, if we combine institutionalist studies with the wider literature on power and knowledge, we find that many women form feminist networks built partly on their experiences of exclusion (Cairney and Rummery, 2018;Woodward, 2004), there is a stronger tendency for women of colour to be abused and threatened in debate (Zevallos, 2017) and erased in intellectual and activism history (Cooper, 2017;Emejulu, 2018), while some forms of knowledge -primarily from the Global South -are marginalised in academic studies and policy debate (Hall and Tandon, 2017;Oliver and Faul, 2018). These imbalances in respect for knowledge claims, and opportunities to communicate or engage, combine with similar types of inequality within the academic profession, in which white men are more likely to be in senior academic positions, published and cited in high 'impact' journals, and submitted to the REF publication and impact process (see, for example, HEFCE, 2015; Williams et al., 2015). To play the REF game without acknowledging these problems is to legitimise and reinforce the inequalities that many of us profess to oppose.

Concluding discussion
On the one hand, the UK's impact agenda is a sincere attempt to provide new incentives and rewards to scholars. The older peer review dominated system tended to reward scientific work that appeals primarily to an academic audience, and the rewards for impact perhaps encourage a change of mind-set in some academics, or provide more reward for academics already invested in social and political impact. In that context, the 'how to' advice is very useful to scholars new to the field, seeking advice on impact, and uninterested in reinventing the wheel or learning primarily from their own mistakes. Many people have engaged in impact activities and their experiences provide a wealth of practical information and reflection on dilemmas.
On the other hand, the written rules of impact often help exacerbate the unwritten rules of professional inequalities. For example, the resources and opportunities to seek impact are not shared equally, and the current system has encouraged Universities to invest primarily in stories of heroic scientists -usually white male professors -overcoming the odds to impact (Dunlop, 2018). In contrast, women and 'those from BAME backgrounds' are the least likely to engage in professionally rewarding impact activities such as giving access to parliaments (Foxen, 2017;Geddes, 2018). More generally, people of colour are under-represented in senior academic positions and therefore have fewer opportunities to engage (Bhopal, 2018;Khan, 2017). In that context, generic 'how to' descriptions of impact activities hide the highly uneven opportunities, incentives, and payoffs.
Some 'how to' advice seems to scratch the surface of the problem, inviting us to communicate clearly or wait for a sufficiently science literate policymaking audience to appear. Or, it helps highlight (unintentionally) the inequalities of opportunity for academics to produce impact, such as when identifying the need to form relationships with policymakers and engage directly and intensely in political debate. Safe 'how to' advice also helps perpetuate a cycle in which (a) enough people know about, and have described, key dilemmas in the academic literature (Hendriks, 2002;Jasanoff and Polsby, 1991;Pielke, 2007), but (b) the messages are crowded out by naïve or normative understandings of research and policy, which leaves (c) each generation of scientists to learn the same lessons through trial-and-error over many years rather than at the beginning of their career.
Regardless of the hand you choose, the policy theory literature helps us make sense of the 'how to' advice in practice. To be a 'policy entrepreneur' is to find out where the action is, learn the rules of the game, form alliances, frame your evidence in relation to the dominant language of policy debate, and respond to socioeconomic context and events which help create windows of opportunity (Cairney, 2018b). However, to be a reflective scholar is to recognise that few entrepreneurs succeed, and relative success results more from societal structures and the policymaking environment than simply from skilful entrepreneurship.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.