WEEKEND READING: A snapshot analysis of UKRI’s Diversity Data

This blog was written by KCL Physics PhD student and current HEPI intern, Bethan Cornell. Bethan is the author of the recent HEPI publication, PhD Life: The Student Experience. Find Bethan on Twitter @CornellBethan.

On Thursday, 16 July 2020, Bethan will be speaking at a free HEPI webinar on PhD students – further details are here.

On 24th June, UK Research and Innovation (UKRI) published harmonised diversity data of applicants to and recipients of Research Council grants over the last five years. Here, I look at what the data says, what we might learn from it and what policies might be put in place as a response.

Gender

The UKRI analysis of Research Council awards broken down by gender states:

In the last five years, PI [principal investigator] applicants who select male as their gender had higher awards rates than those who select female as their gender.

Overall, this is true for all the levels considered in the data (Principal Investigator (PI), Co-Investigator (CIs) and Fellow) but most pronounced for PIs. However, there are a few exceptions to the rule: the Arts and Humanities Research Council (AHRC) has higher award rates for females than males at all levels. Similarly, in the Natural Environment Research Council (NERC), female CIs have higher award rates than male CIs and at the Engineering and Physical Sciences Research Council (EPSRC) the male-female award gap has been closing year-on-year – down to 1% in 2018/19.

The data do not just show a gap in terms of awards given. In almost all Research Councils, female PIs who do win awards win less money than male PIs. The only clear exception being in the Biotechnology and Biological Sciences Research Council (BBSRC), where the amounts given to female PIs are either comparable to or more than those given to male PIs. Overall, across UKRI, males were awarded £32,000 more on average according to the mean and £49,000 on average compared to the median, than females.

When it comes to awarding studentships, there are also gender disparities. The UKRI summary report states:

Approximately 40% of recipients of UKRI studentships in the last five years were female, which is less than the HESA [Higher Education Statistics Agency] estimate of females in the PGR [postgraduate research student] population (49%).

However, it does go on to say the proportion of females awarded grants has risen two percentage points since 2014/15.

It was never in doubt that it is harder to progress in research as a woman than a man and this UKRI data quantifies that fact. It is concerning to see such stark disparities between the genders, especially since these persist at all levels both in terms of the number of successful applications and size of awards given. Nevertheless, it is encouraging that some Research Councils seem to buck the trend or are showing a narrowing gender-divide. This implies some good work is being done to improve the situation for female researchers.

Ethnicity

Unfortunately, in the UKRI data a high proportion of applicants’ ethnicity is unknown, which limits the conclusions that can be drawn. However, the data that are presented show a striking underrepresentation of ethnic minorities in research. White applicants dominate successful applications at all levels considered. In fact, in the Medical Research Council (MRC) and in NERC, the white-ethnic minority award gap is widening. Only one Research Council runs against the trend: AHRC ethnic minority applicants are more successful than white applicants.

The data also show that across UKRI, over the last five years, successful applications from ethnic minority backgrounds were awarded less money than white applicants. When comparing the mean value awarded, ethnic minorities get £106,000 less on average than their white counterparts. This drops to £30,000, however, when comparing the median.

Just like when comparing genders, the data shows that even at the PhD level, there are disparities. There is a clear favouring of white applicants. The UKRI summary report acknowledges ‘the proportion of ethnic minority studentship awardees is less than the PGR population estimate’.

Are things starting to change?

It does seem that the sector is waking up to issues surrounding diversity and inclusion within research. The release of these data coincides with the Black Lives Matter movement, which has sparked responses among researchers. For example, on 10th June, the #ShutDownStem movement encouraged Science researchers to suspend work for a day and think about diversity and inclusion issues in their field or workplace. On 29th June, an open letter, signed by 320 academics was sent to the Executive Chairs of UKRI and all the Research Councils, demanding a change to the PhD recruitment process. The signatories believed:

Many [Doctoral Training Partnerships and Centres for Doctoral Training] are scoring and ranking candidates on the basis of a narrow set of criteria rooted in biased views of ‘excellence’. These criteria implicit ally capture access to opportunity rather than ability and potential.

Having said that, there is a long way to go. As one example, the Institute of Physics announced its Honorary Fellows at the end of June, in which it awarded more to men called Brian than to women, and to no one at all from an ethnic minority background.

Policy recommendations

The UKRI data shows clear room for improvement in terms of diversity and inclusion within academia. However, there is currently an appetite among many researchers to make this happen. A good starting point could be:

1. For Research Councils who do well in terms of diversity and inclusion so share their good practice with those who do not.
2. For institutions to actively promote under-represented groups by identifying them among their staff, encouraging them to apply for fellowships and then carefully supporting them through this process.
3. For UKRI to ensure that all its peer review grant panels are representative of the different demographics in research, ensuring the diversity of researchers is matched by proportionate diversity on the panel.