Science-funding agencies are increasingly creating metascience units to study, using scientific methods, how research is funded, conducted and evaluated. The United States has called for every federal science agency to establish one, while Ireland is also developing a similar team.
The UK Metascience Unit, launched in 2024, was the first team of its kind embedded within a national research-funding system. It is jointly operated by the UK government’s science department and UK Research and Innovation (UKRI). Over its first two years, the team conducted experiments and analysed data to identify more effective ways to fund and support research and innovation.
One trial examined whether emails appealing to potential grant reviewers’ sense of social responsibility would increase acceptance rates. They did not. Another study assessed distributed peer review, in which applicants review one another’s proposals. The approach accelerated grant assessment because the reviewer pool grows with the number of applicants, reducing the need to recruit external reviewers.
The unit is also testing staff desk rejection. Under this process, UKRI employees identify and remove non-competitive applications before peer review, with the aim of reducing pressure on reviewers.
Because the UK team was a first attempt, it had no established model to follow. One of its main conclusions is that metascience units should remain small. The team argues that smaller groups are more likely to challenge existing approaches, while larger teams tend to develop ideas that are already established.
The UK unit has operated with six people, and its experience suggests that this size helped it change direction quickly. After finding that email messages had no effect, the team chose to move on to new interventions rather than refine an unsuccessful approach. Its experience indicates that metascience units need freedom to experiment, resources to support that work and leaders prepared to act on the evidence they produce.
Sources: Nature.