Growing numbers of research funders are adopting randomisation as a fairer mechanism to distribute funds in tightly contested funding rounds. Ken Emond explains why the British Academy has used this mechanism for its Small Research Grants and how in its first three years it has improved outcomes.

Three years ago, the British Academy started a major experiment and became one of the first research funders in the UK to make grant awards using partial randomisation. We have decided to extend the trial for a further three years as the outcomes have been very encouraging.

We’ve been running the partial randomisation trial on our Small Research Grants scheme. One of our core funding programmes, it has been going for over fifty years and is one of our most popular and over-subscribed schemes. Offering grants of up to £10,000 for a very flexible range of funding needs, from travel and fieldwork to research assistance, to the holding of workshops to advance research ideas. The awards are often used as pilot studies or first grant opportunities by early career researchers with limited track records as principal investigators.

The innovative nature of the proposals and awards we make as Small Research Grants felt fitting to trial this experimental and innovative method of selection. This is because we realised that some potentially strong candidates might be disqualifying themselves from applying, feeling that their research would not be considered by the Academy, either because of their personal or institutional characteristics or because it might be considered too risky.

So, back in 2022, as part of our work to become a fairer and more equitable funder, we gathered evidence of how partial randomised allocation might help these applicants and discussed this with our Fellows and other collaborators. This led to the initial three-year trial period until 2025, that we’ll be extending to 2028.

Randomisation of grant awarding works especially well in programmes where there are many more fundable applicants of a roughly equal standard than there is the funding available to support them. It also works very well in programmes where the funder is open to supporting riskier, more innovative projects that might not otherwise be selected through more conventional peer review.

Both of these factors came into play in our Small Research Grants programme. Before the trial was launched, we were receiving over 600 applications per round, about half of which were considered fully suitable for funding. The Academy’s subject co-ordinators were having to spend considerable amounts of time making exceptionally fine-grained decisions between applications of a very high standard once the initial peer review had taken place. There was often very little to choose between candidates and it was hard to say one was better than another or more deserving of funding. Using randomised allocation to make final decisions was an obvious step to reduce this burden and take away any perception of personal bias in such decision-making. Now, all applicants who pass the quality threshold as good enough to be fundable have an equal chance of being randomly selected to be funded.

The results of the trial so far have been especially promising. Many more applicants submitted applications, knowing that as long as their proposal is considered good enough to be funded, their chance of being offered the grant is as high as anyone else and that the possibility of unconscious human bias was removed. The total pool of applicants has risen by about 70%, with over 1,100 applications submitted in the most recent round, compared to around 600 beforehand. In each round since we introduced randomised allocation, around 60% of applicants have passed the initial quality threshold to be entered into the randomised allocation.

Despite the increasing numbers of applicants, the success rate continues to be between 20-30% – well above the norm compared to many funding schemes – thanks to maximising funding from the UK Government’s Department of Science, Innovation and Technology (DSIT) and from our partners, including the Leverhulme Trust, Wellcome and Society for the Advancement of Management Studies, among others. At the end of January 2025, there were around 900 awards in progress in the scheme. 

The pool of applicants and of award-holders has become more diverse, with a notable increase in the number of applicants and awards held by researchers of Asian or Asian British background, and a rise in those of Black or Black British background. Institutional diversification has also been notable with awards going to some institutions either not previously supported by the Academy or only rarely supported, such as University of Abertay, University of Dundee, the Alan Turing Institute and the University of Northampton. The 900 awards in progress at the end of January 2025 included researchers based in 112 different institutions, as well as over 40 independent researchers.

Now that the trial has been extended for a further three years, the next stage of the experiment will look more closely at any changes in the outcomes from the awards. Previously, award-holders have typically produced three publications from each grant and have disseminated findings at six events. On average, for every £1 we invest there is a leveraged return of £5.40 from non-UK government funding of this further research  – and a much higher return if subsequent UK government funding (such as UKRI and Research Council grants) is also included. It will be interesting to compare this with future outcomes from grants funded through random allocation to see the impact.

Many funders across the world are developing a diverse range of experiments in assessment processes with the aim of improving efficiency, fairness and equality of access. We are proud to be blazing a trail in the UK, playing our part towards a fairer and more inclusive research culture. We will continue to be active in sharing our experience to date and in looking for more ways to innovate.

The content generated on this blog is for information purposes only. This Article gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science. Please review our comments policy if you have any concerns on posting a comment below.

Image Credit: Frank Fiedler on Shutterstock.