Open Research Online Rethinking Research Impact through Principles for Fair and Equitable Partnerships

With renewed investment of the UK’s official development assistance (ODA) commitment into research, there is a need to rethink traditional understandings of ‘research impact’. In this article, we argue that impact in ODA-funded research should go beyond translating research findings into practice and policy or implementing research in partnership with research mediators/users. Instead, development agendas of those living and working in the global South, including academics and practitioners, and those working in international non-governmental organisations should influence the research agendas, approaches, and schemes that allocate funding. These stakeholders have profound knowledge of what real-world impact looks like, the types of impact needed, local and national realities, and how complex processes of development impact unfold. Drawing on a programme of research conducted by the Rethinking Research Collaborative, we examine eight principles for ‘fair and equitable research partnerships’ using insights from our individual experiences to offer new thinking on ODA-funded research impact.

countries as its main objective. 5 As such, any research funded through the GCRF must: l Seek to investigate a specific problem or seek a specific outcome which will have an impact on a developing country or countries; l Provide evidence as to why this is a problem for the developing country or countries; l Address the issue identified effectively and efficiently; l Use the strengths of the UK to address the issue, working in collaboration with others as appropriate; l Demonstrate that the research is of an internationally excellent standard; l Identify appropriate pathways to impact to ensure that the developing country benefits from the research. 6 These criteria imply that research must not only be problem (or solution) focused, relating to the 'real-world issues' and contexts of developing countries. It must also be clear from the onset about how it will create impact, in relation to economic development and the welfare of countries on the Development Assistance Committee (DAC) list (DFID 2016). Although how impact is understood in this context is not clarified.
Research impact has been an increasing priority in the UK since the focus of Tony Blair's government on evidence-informed policy in the late 1990s (Nesta/Alliance for Useful Evidence 2016) and the continuing 'impact agenda' for higher education, channelled through the Research Excellence Framework (REF) (HEFCE 2011(HEFCE : 17, 2017Nurse 2015). The emphasis has shifted somewhat away from communication of research outputs to engaging users and mediators in research processes (see Fransman 2018). However, despite new requirements to describe 'pathways to impact' in order to secure public research funding, the focus remains on a relatively simplistic supply-side model of impact (Boswell and Smith 2017) which positions the expertise in academia and tends to attribute impact to the individual academic (Dunlop 2018); rather than recognising the complexity of systems and the difficulties of separating the effect of individual action from systemic effects (Cairney and Oliver 2018).
At the same time, those working as practitioners or policymakers in the international development sector engage with the complexity of 'creating impact' on a daily basis, and acknowledge that impact is dependent on a range of factors and actors (Datta 2012;Green 2016). Whether and how outcomes are achieved not only depends on an organisation's knowledge, skills, and expertise in programme design and delivery, but also on more difficult-to-control issues such as timing, political expediency, individual personalities, and relationships.
For activities to contribute to impact depends on deep contextual understandings, well-developed theories of change (which make the link between vision, outcomes, activities and assumptions) and buy-in from a range of individuals and groups. Such practices also operate within (and are therefore constrained by) national and international policy systems (see Carbonnier and Kontinen 2014;Kok et al. 2017).
The ability of any development activity to create impact is therefore complex and social -it involves an array of actors and implies the importance of a collaborative approach (Georgalakis 2017). In this article, we argue that for ODA-funded research to contribute to realworld impact, we need to ground that research in the experience and current practice of development practitioners, and their knowledge and understanding of what impact is needed and how this might be created. This suggests the need to radically rethink our assumptions in this area. This means that rather than starting with research priorities and the implications of the research design for 'pathways to impact', we need to take the ongoing development activities of practitioners as a starting point, working with, and within, their wider processes of change. This rethinking has implications for how research is understood in relation to other knowledge systems, how research funding is allocated, and how research collaborations are designed, implemented, and supported.
This need to rethink what is understood by (international development) research impact, and how that impact is created is a key concern of the Rethinking Research Collaborative (RRC  Figure 1).
These evolved to place greater emphasis on dialogue, engagement, or collaboration (see Figure 2) while retaining the linear relationship between spheres of research (which involves varying degrees of participation) and the sphere of impact.
A third wave of models began to recognise research impact as part of a more cyclical process of knowledge production and policymaking (see

Research
Policy/practice Impact Engagement/ co-production Figure 3). While 'impact' is located in the sphere of uptake/adaptation/ use, the nature of these complex processes will be affected by the ways in which research agendas have been set and the nature of research production and communication.
More recently, understandings of impact evolved to recognise impact as part of more complex systems of knowledge mobilisation (see Figure 4).
These different models have implications for the way research impact is understood, the types of collaboration that are involved, and how capacity is strengthened.  In the first three models, the focus is on the academic research system, which might or might not invite collaboration from non-academics. In contrast, the knowledge mobilisation cycle at the heart of Figure 4 invites academic research processes to engage with the ongoing knowledge activities of development practitioners (see Hayman et al. 2016;Mougeot 2017), while recognising the effects of the broader national and international knowledge systems in which the cycle is embedded (Bradley 2017;Kok et al. 2017;Lebel and McLean 2018;UKCDS 2017).
We argue that this fourth type of approach is necessary both to fulfil the ambitions of ODA compliance in research impact; and to adhere to the implications of the eight principles developed for fair and equitable research partnerships. This has implications for how research impact is conceived and understood; as well as how such impact can be created. We suggest that the principles, initially developed to support the process of fair and equitable research partnering, can (and indeed should) be applied to give meaning to the concept of research impact. Reflecting on the principles enables us to develop a deeper understanding of the potential for research impact, and how it might be achieved. Responding to these insights suggests a need for different types of relationships between UK-based academics and their research partners; which in turn can enable a more diverse, inclusive, and relevant approach to the production of development knowledge.
3 Eight principles for fair and equitable research partnerships

Background and methodology
In early 2018, UK Research and Innovation (UKRI -a body set up to coordinate work across the seven research councils, Innovate UK, and Research England, and to create an environment to enable research and innovation to flourish) was considering how to strengthen the ODA commitments of the GCRF in response to some criticism from the Independent Commission for Aid Impact (ICAI 2017). 'Fair and equitable' research partnerships were a key element of the GCRF's approach to delivering these commitments (UKCDS 2017), but there was limited shared understanding of what the idea of 'fair and equitable' research partnerships meant in practice.
In response, the RRC proposed some strategic research (involving the collection of qualitative data through interviews, focus group discussions, and written statements) to explore the perspectives and experiences of research 'partners' (i.e. those individuals, organisations, and networks who were not based in UK academic institutions, but had been, or had the potential to be involved in ODA-funded research; these included INGO and research broker organisation staff based in the UK, and academics and civil society based in the global South 7 ). Over a one-month period, the RRC reached out to its extensive networks, with each of the co-investigators targeting partners from their own stakeholder group. Respondents reflected on their experiences as partners in research, focusing on the factors which enabled and constrained their participation. It quickly became clear that in considering fair and equitable research partnerships, it is not enough to look at a specific partnership in isolation; rather, it is necessary to consider the entire research system -from agenda-setting, to research design, implementation, and communication, as well as the mobilisation of knowledge into practice and policy beyond the research.
Building from the research, we held a roundtable event in April, bringing together key representatives from UK-based research funding bodies, GCRF strategic research leads, and representatives from the different partner organisations to discuss our findings. We also carried out a literature review to examine existing resources on partnerships.
The research and discussions led to the development of a set of principles for fair and equitable research partnerships and a set of resource materials, 8 targeted at different stakeholder groups, to support translation of these principles into policy and practice.

Eight principles
The eight principles identified are as follows: 1 Put poverty first.
2 Critically engage with contexts.
4 Adapt and respond.
6 Commit to transparency.
7 Invest in the relationship.
In the following section, we take each principle in turn, and consider its implications for a renewed understanding of research impact. We introduce the principle through a quote from a 'research partner' (collected during our research) which influenced the formation of the principle. We then unpack the meaning of the principle, drawing on an example from practice to identify the implications for an embedded, systems-based approach to research impact. Reflecting on this experience suggests that if research is to have realworld impact, then it is not enough to articulate a research question that appears to address a development challenge; it is not even enough to design a good participatory process, which involves those who are facing the particular development challenge in question. Rather, it is important to locate the research within a wider practice-based theory of change -and to understand how it will 'land' in the broader context. Those designing the research need to have a deep understanding of the actors, processes, and interventions already at play, and of the different dimensions and paces of change; to understand how their research integrates into ongoing work.

Applying the principles to a renewed understanding of research impact
Attention to these issues influences the research questions and who is asking them; the research design and whose voices are included; the research process and who the researchers are; and how creating impact will be considered and actively sought through the process. But beyond this, it suggests a shift at every level of the research system -including rethinking how research agendas are set, and how incentives are integrated into the system. For example, research funders would need to consider how to encourage and enable discussions to take place, so that those involved in supporting practical development work are able to become influential in how research is framed, planned, and implemented, ultimately to create the best possibility for impact throughout. Although contexts shape the possibilities for partnership, and for research, no context is fixed, they are dynamic. In mapping and exploring different aspects of the context -considering who the actors are, what motivations they have, their evidence preferences, and what opportunities and constraints these present, as well as what 'power' means (i.e. who has it, why, and how is it used) -we can start to appreciate the different dynamics which shape research and influence its potential for impact. This can lay the foundations for pragmatic decisions on the extent to which such dynamics are confronted through the research and the partnership, worked around, or merely acknowledged in our research approach and impact planning. More fundamentally, this suggests that a rooted power and political analysis needs to be at the heart of a research-for-impact process, suggesting the importance of recognising diverse skills within any research process (see Principle 5). This means that we need to revisit systems that classify different types of evidence and shape the distinction between 'grey literature' and valid (academic) evidence and knowledge. Key to this is to build more democratic systems of knowledge certification, to decentralise control over peer-reviewed journals and other mechanisms that widen chasms, and to invest in processes to encourage and enable different types of evidence to flow into international development knowledge. For example, through involving non-academics more in research agendasetting and allowing them to take on leadership roles. This will ensure that the knowledge generated from any research is deeper, shared on its own terms with its own framing, voices, and positionality, more closely aligned with the multiple and complex processes of change, and thereby more able to create impact.

Unnecessary controls in the process need to be done away with. There should be an option for an optimum degree of flexibility in the process, and more importantly, space for creativity and innovation… This is because we in the South can see certain things which others cannot see, and therefore, we should have enough liberty and freedom to change course, when necessary. (Academic based in the global South)
Social change is complex and complicated, and rarely follows a linear planned path. Although there is continued pressure from funders and policymakers for those implementing development programmes to clearly identify project outcomes, and follow neatly designed programmes of activity to reach these, there is also recognition that such pathways do not exist. Whether intended activities create the intended impact depends on the individuals involved, the (shifting) context, and broader socioeconomic, political, and environmental dynamics. More progressive development funders are increasingly open to funding interventions that specify a programme objective, but enable flexibility in how that objective is reached, valuing space to reflect, learn, and adapt as programmes are delivered. Much can be learnt from the long-term and relational practices of NGOs which have a deep, engrained understanding of the contexts in which they work, and are adept at responding and adapting practice accordingly. Real-world research can be challenging -the 'study area' interacts with an ongoing process, it is not a test location or a bounded discrete entity or experience. This can make planning and implementing research complex. Taking a responsive and adaptive approach can enable research to embed more deeply in its context and unearth greater possibilities for impact. Social change strategies often include processes of horizon scanning and adapting to shifting contexts and opportunities. This also implies that pathways to impact should have inbuilt flexibility to evolve over time and must be revisited, collectively, throughout the research partnership. The ability to do this will depend on strong partnership relationships with clarity of vision for the research, alongside deep understanding of the dynamic context in which it is taking place. Making a research partnership greater than the sum of its parts means taking time at the outset to explore the knowledges, skills, and experiences that each partner brings; this includes recognising the full range of skills, beyond academic expertise, needed for partnership and for impact. Moreover, if that research is to create development impact, this will also involve exploring the views, perspectives, and aspirations of those whose lives the research is intended to change. It is not enough to understand skills, but also to consider the motivations and expectations of different partners, and each other's institutional contexts.

IDS Bulletin
Being associated with an organisation that uses participatory methods and approaches to sustainable development, I find that participatory research has often emerged as a way of bringing in diverse perspectives. We are frequently approached to facilitate engagement with marginalised groups to bring their voices to the fore and often this engagement is at the core of the research output. Divergent views within the various community groups, between the researchers and the researched as well as among the various sets of researchers, emerge. The struggle is to navigate the complexities of these diverse views and at the same time, do justice to the processes that facilitated the articulation of these views.

(Sowmyaa Bharadwaj)
While researchers are good at research, they may be less capable of creating change strategies, or reading and responding to opportunities to bring about change and impact, than the group of people for whom the impact is intended. Those living in poverty and experiencing discrimination and marginalisation on the other hand, when provided appropriate opportunities, are very capable of designing strategies that have far-reaching and meaningful impact. Thus, researchers need to ensure that such diverse individuals' skills are valued and respected as a foundation to creating impact.
The way that change and impact are understood and prioritised as well as understanding what types of change may be most impactful (for example, is it about those involved acting on new knowledge, or about policy change, or theoretical understanding?) requires a concerted effort. In order to create impact, we therefore need to create space to consider different perspectives on, understandings of, and needs for impact; and value diverse and distinct pathways for the importance they hold for different members of the partnership. The entire grant process should be carried out in a structured, organised, and transparent manner. Aspects like budget and funds disbursal should be free from ambiguities to avoid any conflicts later on. There should also be flexibility in how and where the money flows, to avoid any stakeholder exerting undue rights over research funds. (Academic based in Asia) A code of conduct or a memorandum of understanding are useful ways to make explicit the commitments of each partner in a research partnership. These include administration and budgeting, and the rights of all partners regarding acknowledgement, authorship, intellectual property, and data use. But despite the possibilities offered by mechanisms like these, there are underlying challenges about the meaning of transparency. In Praxis we have had similar struggles, and have learnt that transparency needs to be about the relationships rather than a legal document. For example, when agreeing copyright there is often a long complex exchange with the legal department of the contracting entity. We might be assured of flexibility, but when the work is complete and we seek 'permission' to use the data that we have collected, we are politely refused. Rather than engage in a debate about copyright itself, we are referred to the copyright clause that was signed off on in the contract. It's almost as if transparency is wholeheartedly encouraged by the more powerful and those controlling the finances -and they put in several clauses that showcase transparency, but because they have access to systems they can ensure that they seem benevolent while power hierarchies remain in the status quo.

(Sowmyaa Bharadwaj)
Challenges to transparency affect the wider processes surrounding research -including who feels ownership of the process, which can limit the potential for impact. Partners in a research relationship, in embracing the commitment to transparency, should ensure that they find ways to recognise and question these inbuilt power hierarchies rather than perpetuating them. It is only by considering the wider relationship and behaviours that transparency can begin to operate as a value rather than a transaction. In this way, it can contribute to creating impact within the specific partnership relationship and related research, and beyond to the wider research system, to open up space for discussions on fairness and equity, and to consider how impact is attributed and owned. Strong and sustained relationships are vital for impact, but relationship building is a less tangible aspect of research and seldom funded. And yet, sustained relationships are essential for bridging knowledges, creating common languages, and facilitating impact pathways. Funders could help by earmarking funds for the relationship before, during, and after projects. Partners might also consider the need to work beyond the templated and inflexible contractual dotted line framework and join efforts. This would help to avoid reducing partnerships to the commitments of individual people, by investing in strategies to ensure that the wider organisation benefits from, and has an interest in, sustaining the partnership. Given the long timescale of development impact, such long-lasting relationships are key to enabling sustainable change.
Research partnerships are important because they enable us to work with people that we would not usually work with, and in ways that we do not usually work. This can give us access to new skills, perspectives, understandings, and knowledges; and it can mean that together we can create opportunities for synergies and deeper understandings. Learning is about mindset and openness. Within a research partnership, it is about valuing diversity. If the partnership is to become a site for transformation, it is necessary to deliberately create spaces to enable learning. It is also important to recognise that learning is not always comfortable and can frequently be emotional -while on the one hand we might be learning new skills, we might also be learning about ourselves, our assumptions, and behaviour. Translating that learning into action requires bravery and honesty. Without mechanisms to capture learning and channel it back into onward organisational strategies, the benefits of individual and collective learning will always be limited. culture shift to recognising and embracing the learning from potential failure, acknowledging that while projects might struggle in certain ways, learning from this can be key to enabling impact. This includes a recognition that impact may come in a different form from that initially envisaged. Table 1 summarises the contribution made by each of the principles to our proposed understanding of impact.

Conclusions
By reflecting on and giving meaning to a set of principles, initially designed to support 'fair and equitable partnerships', we can delve deeper into understanding the role partnerships can play in contributing to more impactful research. Our experience and our analysis suggest that there are multiple factors that need to be considered if research is to become properly embedded in systems of development and change.
Theory on research impact has evolved in recent years. What started with assumptions of linear transfers -linking knowledge and impact directly -have become more complex, recognising both the need for co-production during research processes and considerations of cyclical models of research uptake, adaptation, and use. However, to date, this theory has engaged with impact from the perspective of the academic research system. By drawing on the eight principles for fair and equitable partnerships, generated from a 'partners' perspective', and sharing specific experiences we argue that if research is to be truly impactful, it needs to take its direction from the ongoing development activities of those working actively to bring about social change. By taking practice as the starting point, and appreciating the complex pathways to creating change, it is more likely that research will be able to contribute to real-world impact. This shifts the way research impact is understood, to ensure that it responds more closely to the knowledge and experience of those working directly to challenge poverty and contribute to social justice.
Operationalising such an embedded approach to impact has implications at different levels, for different actors involved.
Academics based in the global North need to be properly incentivised and supported to develop deeper impact models. This includes establishing the skills and capabilities they need to be able to properly engage with the understanding, knowledge, and experience of those outside the Northern university environment. It also means ensuring that a process which encourages the reporting of impact also takes into account the complexity of pathways, and focuses on valuing contribution rather than attribution, recognising the range of dynamics that affect the potential of any piece of research to create impact. Likewise, academics in the global South need to be encouraged to identify where collaboration with Northern-based academics brings value, and where collaboration with civil society practitioners may be more appropriate.
Understanding impact as complex, multifaceted, adaptive, cyclical, and long term has implications for research governance and agendasetting. Research funders in the global North will need to develop new funding models with greater representation of a diversity of development actors in funding decision-making. New investment must be made to ensure adequate time for mapping impact contexts and actors, to respond to complex development impact pathways, and to build strong, open relationships which recognise and combine different knowledges, creating spaces and systems for learning throughout.
Finally, civil society practitioners based in the global North and South also need to adapt: this includes recognising and asserting the skills and understanding that they have developed through their development interventions and applying these in the field of development research. But it also requires a deeper understanding of what research can and cannot bring into their wider impact strategies. Such practitioners need to understand when research might be useful, what research might be useful, and what skills and relationships are needed to be able to do it.
There are many good reasons for practitioners to engage in research as part of their development interventions; for example: l To understand more about an issue observed as affecting programming and thereby to sharpen a programmatic intervention/ building a better response to a development challenge; l To deepen understanding or gather evidence on an issue which may inform policy development, or be used within advocacy and campaigning action; l To support reflection on and development of internal organisational practice, including how knowledge and evidence are understood within the organisation; l To capture learning and deepen understanding of how work contributes to change, to strengthen practice, to secure funding, or influence the practice of others; and l As a way of exploring, articulating, gathering, and documenting, from the perspective of those involved in programme work, to enable their voices, understanding, and sense-making to inform wider debates and dialogue on specific development issues.
But for research to have impact in any of these cases, it is crucial that the practitioner situates the research within their broader change strategy and invites others to collaborate within their process. This requires that all those involved acknowledge their skills and capabilities, positionalities, and motivations, to work together to enable such responsive embedded research, being honest about both the potential and limitations it offers. By building research agendas and focusing research design in this way, those involved in scoping, funding, and delivering ODA-funded research can create the potential for ODA excellence and enable research to play an active role in poverty eradication.