Collecting and analysing diversity data

EPSRC data at a glance

The diagram below gives a snapshot of the diversity data from the 2016/17 financial year of the approximate size of the engineering and physical science research community (based on data obtained by the Higher Education Statistics agency (HESA)) and the number of people supported by EPSRC. The diagram also shows the number of individuals that applied to EPSRC during 2018/19 and the number of individual subsequently awarded funding. This data includes all individual investigators across schemes and an individual is only counted once regardless of whether they are an investigator on multiple grants (View a PDF version of the below graphic).

Diversity data

Diversity data on applications and success rates

EPSRC publishes diversity data on success rates which include grant and fellowship applications, alongside information on Co-Investigators and also student population data. The fourth annual publication is available on the UKRI website.

Key points from EPSRC's diversity data are:

  • For Principle Investigators, success rates by number became equal by gender in 2015/16 and have remained within 1.5% since then. Data for co-investigators fluctuates.
  • For the fourth consecutive year, women have higher success rates than men in our fellowship scheme.
  • Applications by age are consistent with the HESA data, with a slight under representation of under 30s as Principal Investigators, however our fellowship scheme has a younger age profile. There is no discernible pattern in the age-related success rate data
  • Around three quarters of applicants describe themselves as White, the number of Principal Investigators describing themselves as Asian/Black/Chinese/Mixed/Other has risen by to 18% and Co-Investigators 16%.
  • Success Rates for Principal Investigator who describe themselves as Asian/Black/Chinese/Mixed/Other are consistently lower than those who disclose that they are White.
  • For the first time since 2011/12 the success rate of fellowship applicants describing themselves as Asian/Black/Chinese/Mixed/Other (37.0%) is higher than those describing themselves as white (27.4%).
  • A proportion (7%) chose not to disclose their ethnicity, this is significantly higher in our student population (25%).

This analysis looks at grants by number of applications. We are currently working to extend our data analysis to look at grants by value.

Diversity data related to peer review

We are now publishing further diversity data to highlight the progress made to date to improve diversity in our peer review process, and the opportunities to continue to work with our communities to make further improvements. The data was published in response to a growing community interest in this information. The data covers the 2011-2017 period and includes diversity data by gender, age and ethnicity for the:

  • The EPSRC Peer Review College
  • The reviewer population that were invited to review proposals
  • The reviewer population that submitted a useable review
  • EPSRC Internal and External Panel Chairs  
  • EPSRC Internal and External Panel membership.

View the full data report. Key points from the data are:

  • The percentage of female Peer Review College members has increased to 16% in 2018/19. The extent of the increase in female reviewers is perhaps more evident when looking at the number of female College members which has doubled from 2011/12 levels.
  • Comparison with HESA data indicates that there is a slight over representation in the Peer Review College in the later age categories (50-59, 65+) and under representation in the earlier categories (0-29, 30-39).
  • The data shows progress towards achieving our mixed gender panel policy aspirations, rising from 13.8% female in 2013/14 to 30.3% in 2018/19.
  • The number of Panel Members describing themselves as Asian/Black/Chinese/Mixed/Other has risen from 4.7% in 2011/12 to 8.1% in 2018/19, although this remains lower than the estimated academic population.
  • Data on Panel Chairs shows a significant increase in female representation, from 13.3% in 2012/13 to 30.8% in 2018/19.

We are not publishing disability data for these populations at present. The numbers disclosing a disability are small, and a significant number choose not to disclose or are unknown. We continue to monitor this data internally.