Our approach to working with the UK science 'impact' agenda

For a while now in the UK and elsewhere, there has been a move to encourage academic science researchers to consider the broader implications of their work, and think about how they are going to involve people from outside academia in their research.

This ‘impact’ agenda is pushed by most major research funders, and is an increasingly major part of how universities are assessed in the Research Excellence Framework (this assessment determines how much funding universities receive in the UK).

We (FoAM Kernow) were recently asked by one of the UK research councils to write about how we collaborate with university academics to build ‘impact’ into research projects, partly to provide inspiration for the researchers but also to address some common misconceptions.

Many research funders now require all their grant applicants to write a ‘Pathways to Impact’ statement (in the lingo of the UK Research and Innovation which is the Government funding body for research, or ‘Public Engagement’ statement (example jargon from the Wellcome Trust). Some funders like the EU will expect this type of work to be much more deeply embedded within the main research plan, rather than necessarily splitting it into a distinct section of the application.

But - there’s quite a bit of misunderstanding within universities about what counts as ‘impact’, and most academics never receive any serious support or training in this, which means rumours abound within the research community, and the funders don’t always get what they’re after. So, nobody is quite as happy as they could be.

One common misunderstanding that we come across from academics is that what they refer to as ‘public engagement’ (basically anything where people from outside academia are told about or involved in research) does not count as ‘impact’. There is decent evidence to counter this from the National Co-ordinating Centre for Public Engagement, where they analysed the Impact Case Studies submitted to the REF:

"There was no significant difference in the scores awarded to case studies featuring mentions of public engagement compared with those that don’t: anecdotally, there was nervousness in the sector that public engagement would be valued less highly than other types of engagement. This finding challenges that assumption."

Because we work directly with such a wide mix of academics and funders, we often end up as the go-between. We frequently hear academics grumble that they have to do impact statements at all, and also hear funders lament the relatively unimaginative ‘Pathways to Impact’ they receive, with ‘public engagement’ usually relegated to a talk in a school or at a local community centre. This type of performative and didactic approach, where the academic simply imparts information to their chosen audience, is beginning to be deemed inadequate.

This seems like a relatively straightforward problem to fix. Funders are beginning to write more detailed requirements about what they expect, and starting to bring specialists in public engagement onto their funding decision panels. There are moves to modify funding regulations to make it easier for academic researchers to work with external organisations that have more, or different experience (indeed this is already straightforward with funders such as the EU, AHRC and Wellcome Trust). This approach means the academics can learn directly from experts, and also don’t have to do all the work themselves.

Another common issue is that researchers are wary of how much they cost into grants for ‘impact’, and this means that often the work is under-budgeted or significantly less ambitious than it could be. Indeed, funders have expressed to us that they have difficulties persuading researchers to cost in this work in any serious way. To give some concrete numbers, we have costed up impact plans at £15-30k for UKRI grants, £40k for Wellcome Trust, and £80k+ for EU research grants. The amounts allocated to pathways to impact should generally be proportionate to the overall size of the research grant, and 10-20% is about right depending on the scope of the work and how deeply integrated it is with the main research outcomes. We have experienced hesitation from academics that this level of funding is too high, and researchers often believe that reducing the total might make the funders more likely to bite. However – to date, we have never had a proposal where the funding awarded has been reduced from what we asked, and indeed from our experience the likelihood of a successful proposal appears to be proportional to the scale and ambition of the impact plan.

Our approach with this type of project is always to first see if there’s a way that the ‘impact/engagement’ component can be designed in a way that means the research itself benefits. The gold standard for us is a situation where people from outside academia are given an opportunity to directly influence the research questions or culture, or be enabled to do their own research. We try to push projects up levels in our hierarchy of people’s involvement in scientific research (we’ve adapted this from various similar hierarchies e.g. see Prof. Muki Hacklay’s blog):

[

LOWEST LEVEL

Receive education (e.g. exhibitions, talks)

Contribute processing power (e.g. BIOINC)

Contribute data (crowdsourcing e.g. Zooniverse, wildlife surveys)

Problem solving (e.g. foldit)

Question raising (e.g. Dutch science shops)

Drive own research (e.g. DIYbio)

Co-create new scientific culture (e.g. AccessLab, Biohackspaces)

HIGHEST LEVEL

]

Some example projects to give a flavour of what is possible

Viruscraft – This is a Wellcome Trust funded research project combining tangible interfaces, craft and computer games. It was written into a fellowship application as a collaboration for the public engagement component of the project. Here we work with evolutionary biologist Dr. Ben Longdon who is looking at what determines the ability of a virus to infect some hosts but not others. We have built a large wooden virus, rigged with sensors to detect wooden shapes that are plugged into the surface. In this way, players ‘evolve’ their virus, infecting and jumping between host species in a screen-based game world. The game runs on a host-pathogen evolution simulation, so the players are directly interacting with the research data, and can run experiments by setting up the hosts’ world, and evolving their virus in different ways. The development of the tangible interface is (publishable) research in its own right. Co-development workshops were used from the very start to make sure a broad range of people could be deeply involved in the project.

Tanglebots – This was originally part of an AHRC funded project, in collaboration with Alex McLean, Ellen Harlizius-Klück and Lovebytes. Tanglebots is a workshop format that we developed for children and their families, forming a messy introduction into weaving, robotics and coding. We also later received some ESRC Public Engagement funding, run the Tanglebots workshop for children with autism, in collaboration with the Cultural Minds research group at the University of Exeter. Tanglebots are prototype/failed weaving robots (weaving is quite hard, so we start with tangles and see what patterns emerge). We combine harvesting components from e-waste toys (motors, gears, electronics), and using visual programming (scratch) to control these components, to create robots that make tangles. In this case the participants took the role of researchers, and helped develop ideas for how the research could progress. These workshops directly informed our own research, leading to a new ERC fellowship, and also informed the cognition research of the Cultural Minds group.

Cricket Tales – This is a NERC funded project, where the ‘impact/engagement’ takes the form of a citizen science system developed in collaboration between FoAM and the Wild Crickets researchers. By tagging events in the cricket CCTV videos, players contribute directly to research which will determine whether crickets have individual personalities. At the end of each game play, the latest results are displayed, including the data just contributed. This means the players are the first to see the research results, before the academics – arguably the main appeal of doing scientific research is being the first to discover a new thing, so this puts the players into that role. The system has been developed using two separately funded NERC ‘Pathways to Impact’, and will go live in 2019 as a long-term installation at the Eden Project in Cornwall. Other similar games we have made attract 100-200 players/day in this setting, from all walks of life.

AccessLab – This was a NERC, British Science Association and FEAST funded project, where the bulk of the funding came from the NERC public engagement budget. It is a novel workshop format where academic science researchers are partnered with participants from other sectors to co-research a topic of use/interest to them. Skills in finding and judging information are passed on, meaning the participants leave with the ability to find, access and use scientific research as and when they need to. The workshop format is highly suitable for inclusion in future Pathways to Impact applications.

For more inspiration, other examples of projects that were either funded through impact/engagement routes, or fit well with the impact and engagement agendas include: Sonic Kayaks, Dazzlebug, Greenspace Voxels, the Farm Crap App, Malaria minigames, and the upcoming Behaving Genes.

Advice to funders

One aspect to the science research funding process that we feel is lacking is that little or no formal consideration is given to ‘self-evaluation’. In the arts and cultural sectors, grant applicants are often asked to write a section on how they will notice and evaluate what goes right/wrong with a project, and how it might be improved – indeed it is common to cost in observers who will watch and be present during the research, offering independent feedback to the researchers and publishing their own anthropological research derived from these discoveries. It would be interesting to see whether science funders catch on to the value in this approach, and encourage their applicants to do the same. In the meantime, it is possible to achieve a level of similar self-reflection by encouraging science researchers to aim to publish their ‘impact/engagment’ work in its own right – this approach also fits well with the requirements already placed on researchers, and so may be more palatable than implementing yet more new requirements.