The report, commissioned by the Engineering and Physical Sciences Research Council (EPSRC), was delivered independently by the Royal Statistical Society (RSS) with input from the Alan Turing Institute.
Previous quantitative and qualitative evidence of under-representation and differences in award rates by gender, ethnicity, age and disability, together with input from community dialogues, was used to shape EPSRC’s equality, diversity and inclusion (EDI) action plan.
This new exploratory analysis by the RSS uses their state-of-the-art expertise in data analysis applied to a large pseudonymised nine-year dataset from EPSRC of:
- grant applications and awards
- reviewers’ reports
- panel outcomes
- the personal characteristics of applicants, awardees, panel members and reviewers
It provides a major step forward in our understanding of factors such as host institution, age, sex, and ethnicity, including looking at combinations of these, and their relationship to grant outcomes and participation in application and assessment processes for EPSRC funding opportunities.
New findings
Examples of new findings from the report
While the award rate for minority ethnic researchers is lower overall, looking at the modelled intersection of ethnicity with age, there is less difference in predicted award rates between minority ethnic and white researchers aged under 36 than those in the 36 to 55 and 55 and over age brackets.
When applicants were asked about expectations of bias, institutional prestige was cited as a common concern. However, this was not evidenced in the analysis of outcomes by institution.
Looking at the interactions between the characteristics of the panel and the applicants, in interviews, having any female panellists present is linked to a higher ranking for female applicants.
Exploring the effect of applicant and reviewer characteristics on review scores, the largest effect is higher scoring from reviewers nominated by the applicant.
Full details of these and many other results are summarised in the executive summary and detailed fully in the technical report.
We have opened a Have your say survey for the research community to feedback ideas and reflections on the report.
Fostering inclusion and diversity
Professor Charlotte Deane, EPSRC Executive Chair said:
Working in research is one of the best jobs in the world. We want to attract people with a wide range of perspectives and ideas so that everyone has the chance to experience it.
By bringing new approaches and solutions we can help fulfil the potential of research and innovation in improving people’s lives.
The findings of this report present us with valuable new insights to explore future actions to add to the initiatives we are already undertaking as part of our EDI action plan.
We welcome ideas and suggestions from within our communities to help us embed inclusive practices and increase diversity across our portfolio. Only together, can we create a more inclusive and innovative research environment.
Recommendations and areas to be explored
The report includes recommendations about areas that would benefit from being explored in future data analysis projects. EPSRC welcomes these recommendations and, ahead of further discussion on the report’s findings, is particularly considering:
- furthering work to understand the perception of bias. The qualitative work carried out by the Alan Turing Institute identified bias due to institutional prestige as the main reason for perceived bias among applicants surveyed
- continuing current work to take a qualitative approach to understanding potential biases for people with disabilities and accessibility challenges. Due to the small numbers of individuals in the data with known disabilities, the report was unable to draw firm conclusions about this group
- expanding the analysis to explore the whole team of investigators in a project and understanding how the network effects of their team and organisation may impact application success
Next steps
The report findings related to review quality and how these impact outcomes will inform the ongoing design of UK Research and Innovation’s assessment processes.
The results from the Have your say survey, along with the report’s findings, will inform the future direction of EPSRC’s EDI activities and the update of its EDI action plan.
Significant step forward for transparency
Dr Sarah Cumbers, RSS CEO said:
This work represents a significant step forward for transparency in the funding process of scientific bodies, with the EPSRC being the first organisation of its kind to openly share its data with an external body for independent review.
With the evidence from this data-driven approach, the EPSRC now better understand the barriers some groups face in obtaining funding. We hope that other funding bodies will use these findings to better tailor policies around equality, diversity and inclusion and ensure that our research landscape is diverse in its opinions, ideas and solutions.
Dr Anjali Mazumder, the Turing’s AI and Justice and Human Rights Theme Lead said:
We commend EPSRC for commissioning this review which makes an important contribution to understanding potential biases within the research and innovation grant funding ecosystem. The findings lay the groundwork for further research and interventions to ensure a transparent, fair and trusted process which mitigates bias and ensures that the best research secures funding.
EPSRC’s commitment to EDI
EPSRC funds a diversity of people and ideas to achieve the best research and innovation for the UK.
By embedding EDI in its practices and its supported research and training grants, the impact of research and innovation can be maximised to involve and benefit all parts of society to create an environment where everyone can contribute their best ideas.
The findings from the report builds upon EPSRC’s EDI activities, including:
- piloting the role of a Panel Process Advisor on EPSRC peer review panels. A role carried out by a trained member of the research community with responsibility to ensure the published peer review approach is followed
- encouraging self-nominations to the Peer Review College to improve representation across currently underrepresented groups
- operating a mixed ender panel policy since 2016, which has increased the representation of females on interview panels from 18% to 33%
- facilitating a series of regional roundtables with university colleagues to understand the role university selection processes have on EPSRC’s portfolio with the aim of sharing good selection practices and creating communities of practice