This article is more than 1 year old
RSS has taken a look at biases in EPSRC peer review processes
The Engineering and Physical Sciences Research Council (EPSRC) commissioned the Royal Statistical Society (RSS) to carry out a study into whether there are “demographic disparities in its peer review process and awards, including potential effects due to intersectionality.”
RSS found that “The odds of ethnic minority researchers being successful in a funding application are roughly 32% lower than for white researchers.” An important contextual finding is that the disparity in award is larger between older applicants and there is a higher predicted success rate for having UK as a nationality as a white applicant but this is “considerably weaker for ethnic minority applicants.”
Women applicants are more likely to be successful in funding applications than men. There is little difference in success rates for research grant applications but women are around 80 per cent more likely to be successful in fellowship applications. Overall, the odds of success for women are around 13 per cent higher than men. The discrepancy between the enormous fellowship success and the moderate overall difference is because, proportionately, far more women apply for research grants than fellowships.
In the cases of ethnicity white applicants tend to be awarded larger awards and in the case of sex men tend to be awarded more than women. The analysis suggests this isn’t because panels are reducing the amount of funding distributed but because of the amount of funding that is being requested in the first place. There is a wider question on how individuals are supported to develop proposals and select awards to apply for but this is not within the scope of this specific study.
Ethnic minority applicants were the most likely to perceive bias but the most common reason applicants believed there were biases was due to the institution. There is also evidence that there is a mismatch between the sentiments expressed in panels and the scores awarded to applicants which leads to evidence that “ethnic minority researchers are not achieving the same level of application success as white researchers and a complicated picture around the effect of researchers’ sex on outcomes.”
The number of women as a proportion of the applicant pool has roughly tracked the HESA teaching and research population. However, there are significant discrepancies in the growing size of the ethnic minority teaching and research population and the number of applicants for EPSRC grants.
The use of mixed sex panels seems to have a significant impact on the funding application process. RSS found that “female applicants were ranked 8.5% higher than male applicants when at least once panellist was female but 7.9% lower than males when this was not the case.” RSS recommends that there should be more research on the impacts of previous success on subsequent success of applicants, the impact of diverse panels on awards, and qualitative research into understanding bias relating to disability.
The crucial finding is that there is “some evidence that ethnic minority researchers in engineering, mathematical and physical sciences don’t get the same level of funding as white researchers.” This report doesn’t look at how these challenges may be addressed but there is some tentative evidence that mixed panels, for women, have an impact. The questions it opens is which staff are supported to apply for funding, whether perceptions of excellence unduly leans toward the replication of existing work by existing researchers, and the extent to which some of these challenges are exacerbated by the structure of research funding beyond the composition of any panels.