Go back

Randomising funding ‘could boost breakthrough research’

 Image: Chris Parr for Research Professional News

Arma 2024: Dsit metascience experts look at ways to make funding less risk-averse

More randomisation in the allocation of research funding could shake up the kind of projects that get money, and shift processes away from overly conservative peer review, according to a government adviser.

Jack Leahy, senior metascience adviser at the Department for Science, Innovation and Technology (Dsit), said his team was looking at a range of ways in which the research funding model operates. The team is undertaking a range of experiments to see how metascience—the use of scientific methodologies to analyse and improve the way research operates—might be applied to UK research funding.

“Research funders all around the world often say we really welcome high-risk, high-reward breakthrough research that pushes the boundaries of science and pushes the frontiers,” he told the Association of Research Managers and Administrators’ annual conference in Brighton.

“There is a concern that the peer review process itself has a conservatism bias, and that the really out-there proposals aren’t getting through peer review.”

He said the Dsit team was exploring the option of intervening in the process to remind people that a certain funder “really welcomes…high-risk, high-reward research”. “We want to see if that has an outcome on review scores,” he said.

The ‘grey zone’

There would always be more high-quality funding applications than funding available, Leahy said, and there were a number of options available to drive the awarding process—such as introducing more randomisation in an attempt to reduce bias.

Randomisation, he said, would only be used to allocate funding after scrutiny of core assessment criteria and scoring by those with suitable expertise, but could help redirect money towards applications in the ‘grey zone’—where the available budget is exhausted before projects deemed to be of suitable quality for funding have received a grant.

‘Communities not linked up well’

Ben Steyn, head of metascience at Dsit, told the conference that there were metascientists who were academics. along with others involved in policy practice and innovation.

“My issue is that these communities—the academics and the community of practice—are not linked up very well,” he said, comparing to the situation between “pure research on one side and [the development of] commercialised products on the other”.

“We have the same issue. We have an academic community who are generating insights [but] these insights aren’t being picked up and turned into policy [by funders or universities]”, Steyn said.

A version of this article also appeared in Research Fortnight