Written Evidence Submitted by The Inclusion Group for Equity in Research in STEMM (TIGERS)

(DIV0071)

Executive summary

There is extensive evidence that women, people of colour and disabled people are under-represented in the academic STEM research sector. Career progression in this sector is driven and controlled by research grants, fellowships and studentships from UK Research and Innovation (UKRI) and other publicly or charitably funded bodies. Inequity in the provision of this funding limits access to and progression along STEMM career pathways, and innovation suffers as a consequence. Data from UKRI and others reveal startling inequities in research funding facing scientists from under-represented groups. In order to reduce these inequities, and hence to increase diversity in STEM, TIGERS recommend:

●        Financial incentives to encourage academic institutions to increase the proportion of funding applications from groups which are under-represented in the applicant pool.

●        A requirement that publicly funded grant-giving bodies publish comprehensive quantitative and qualitative data related to the diversity of grant applicants and awardees.

●        An independent body or auditor to be set up to scrutinise policies and practices of research funders with regard to meeting their public sector equality duty.

●        Creation of a new “Universal Basic Research Income”, available to all permanent academic staff members within a framework which incentivises the recruitment of minoritised researchers.

●        Implementation of international best practice in funding policy for improving diversity from the outset at the proposed Advanced Research and Invention Agency.

Introduction

The Inclusion Group for Equity in Research in STEMM (TIGERS) is an inclusive network of students and professionals in science, technology, engineering, maths and medicine (STEMM) which leads and engages with campaigns that aim to make STEMM research more equitable and supports members in their related campaigning activities.

A major focus of TIGERS’ activities is campaigning to improve the fairness of processes and policies used in the allocation of funding for STEMM research. Research funding affects diversity in STEM both directly and indirectly: An example of the direct impact of funding processes is the allocation of personal fellowships which pay a researchers’ salary and often provide a key step on the career ladder. Indirect impacts are manifold and include the fact that an inability to access funding can be a driving force for minoritised academic researchers to leave their chosen field. We thus address the questions posed by the House of Commons Science and Technology Committee Inquiry through the lens of research funding equity. We note that our evidence submission here has significant commonalities with a previous evidence submission[1] to the All-Party Parliamentary Group (APPG) on Diversity and Inclusion in STEM inquiry into Equity in the UK STEM Workforce. The questions addressed by the current inquiry have been asked before and there is ample evidence of inequity available. What is needed is not further inquiries but concerted action to address inequity.

1. The nature or extent to which women, ethnic minorities, people with disabilities and those from disadvantaged socioeconomic backgrounds are underrepresented in STEM in academia

1.1 This evidence submission addresses diversity in STEM in academia only. In this context, STEM workers include (but are not limited to) post-graduate students, post-doctoral researchers, technicians, professional services staff and academics. We address the extent of underrepresentation in academia via data from Advance HE and the Higher Education Statistics Agency (HESA). The Higher Education Statistics Agency (HESA) provides data on the wider UK HE sector without differentiating between subjects for demographic variables other than sex. Advance HE offers subject specific breakdown of gender and ethnicity, but not disability, socioeconomic background or LGBTQ+ identity. It should be noted that the Advance HE data are based on demographic information from the universities who submit applications for Athena Swan awards rather than the whole sector.

1.2 Table 1 (below) presents a breakdown of demographic information of UK staff working on academic contracts, based on available data from HESA[2] and Advance HE (*)[3] for the academic year 2018/19. All numbers are rounded to the nearest 5 or 0. Discrepancies in ethnicity and disability are due to a small number of registered staff with missing relevant data.

Table 1: Demographic information of UK staff working on academic contracts, comparing the STEM workforce with other subjects.

1.3 These data illustrate that women, BAME people and disabled people are under-represented in the academic STEM research workforce relative to the general population. (~47% of the overall UK workforce are women; ~13% are BAME; ~14% are disabled[4]).

1.4 We also note that the numbers in Table 1 conceal significantly greater under-representation of women[5], BAME people[6] and disabled people[7] at senior levels in UK academic STEM, particularly among the professoriate.

1.5 There is a lack of data on the differential experience of specific minoritised ethnic groups. Data tend to be rolled together under the heading of “ethnic minority” or “BAME” (Black, Asian and Minority Ethnic). Moreover, there is very little intersectional data to address the experience of those who embody more than one minoritised protected characteristic, for example Black women (see 5.2).

1.6 We note that the terms of reference of this inquiry do not include the underrepresentation of LGBTQ+ scientists, but will nonetheless include this group in our comments throughout, both because addressing the broad scope of diversity issues is core to TIGERS mission, and because the intersection of LGBTQ+ identities with gender, race and disability leads to further marginalisation

2. The reasons why these groups are underrepresented - addressed through the lens of research funding.

2.1 Gender disparities: Recent data releases from UKRI[8] have indicated that award rates (the proportion of applications which lead to an award) for women who apply for research grants are consistently and statistically significantly lower than the award rates for men. Women also consistently receive smaller grants on average than men. Recently published data from Engineering and Physical Sciences Research Council (EPSRC)[9] showed that for grant values in excess of £2.5 million, the award rate for men is about 40% higher than that for women, whilst men’s advantage rises to almost a factor of 2 for grants in excess of £10 million.

2.2 Ethnicity disparities: There are even greater differences in award rates between white principal investigators and those from minority ethnic backgrounds. Whilst white applicants experience an award rate in applying for UKRI grant funding of 27%, this rate is only 17% for those from ethnic minorities. The average amount of those grants that ethnic minority applicants do receive is almost 20% lower than those received by white investigators[10]. Up until very recently, UKRI data has only addressed “white” and “ethnic minority” categories, thus failing to account for the wide range of experiences and outcomes aggregated under the “ethnic minority” label. A very recent analysis by UKRI has, for the first time, disaggregated grant award rates for different non-white ethnic minority groups[11], and highlighted quite extraordinarily low rates of grant awards to Black researchers: less than 1% of Fellowship recipients are Black and less than 1% of Principal Investigators on funded grants are Black (percentages of funded awards in each case). This analysis also shows that the inequities in funding experienced by Black[12] researchers, and also those from other minoritised ethnic groups, are present from the earliest stages of the research funding pipeline: PhD studentships.

2.3 Disability disparities: Significant disparities also exist in rates of funding received by disabled and non-disabled researchers. While 4.65% of academics identify as disabled (HESA 2019/20[13]), only 1% of applications to UKRI are from disabled researchers. The award rate for disabled researchers is 4% lower than for non-disabled researchers and they are consistently awarded less funding. (In 2018/19, the size of grants awarded to disabled researchers were on average less than half of that awarded to non-disabled researchers).

2.4 Intersectionality: Whilst the available data reveal very significant disparities in funding for women, Black researchers and disabled researchers, they fail to provide sufficient granularity to allow us to examine the experience of those at the intersections of these characteristics. The number of Black women professors in the UK is extremely low[14] (only 0.67% of UK Professors are Black women) , and the proportion of those Black women professors who work in STEM is still lower. We do not know how those women fare in our funding systems; their experiences are invisible in the data. (See 5.2). This lack of data is further compounded by absence of data on additional characteristics which also intersect with gender, ethnicity, and disability.

2.5 Other disparities: Additionally, other aspects of diversity are ignored in UKRI analyses. For example, a report[15] by the Institute of Physics, Royal Astronomical Society and Royal Society of Chemistry on the experience of LGBTQ+ people in STEM reveals that many LGBTQ+ researchers have considered leaving their workplace because of the climate of discrimination towards LGBTQ+ people. Surveys performed in the USA suggest that these issues directly impact LGBTQ+ scientists’ access to funding and employment opportunities[16]. However, there are no detailed data which examine how the operation of UK funding systems impacts LGBTQ+ scientists, despite evidence that the LGBTQ+ community is keen to engage with such data gathering[17]. Equally, there is very little analysis available on the impact of socioeconomic background on the probability of achieving UKRI funding, although there is a significant body of evidence[18] that successful academics in the USA are disproportionately likely to come from well-off backgrounds. This is consistent with the current state of the talent pipeline in the UK, which sees only 12% of STEM undergraduates and 7% of STEM research students coming from the least advantaged quintile in terms of socioeconomic background[19]. (See 5.2).

2.6 Barriers: TIGERS have researched the barriers[20] which current funding processes and policies place in the way of minoritised scientists, focussing on the experiences of women, Black and minority ethnic researchers, disabled researchers and LGBTQ+ researchers. Across all the protected characteristics we have investigated, some common themes emerge that can and should be addressed through government-led policy changes:

2.6.1 Funders often require Universities to perform internal selection to limit the number of grant proposals which are submitted. Often, even when such internal selection is not a requirement, Universities choose to operate internal selection for reasons they deem strategic. This internal selection, which constitutes institutional gatekeeping, is usually carried out with no monitoring of applicant diversity, and minoritised applicants report that they are prevented from submitting funding applications. Internal selection often also places a heavy emphasis on cost, which negatively impacts disabled researchers who require extra costs for reasonable adjustments. A less direct (but pernicious) variant of this problem is seen when minoritised applicants receive less support from University services in developing their applications than more privileged colleagues (typically senior white men). (See 5.1).

2.6.2 Bias (conscious or unconscious) against minoritised groups is an overarching concern in both peer review and panel ranking of research proposals, but when complaints are made funders are often unresponsive, dismissive or display a worrying ignorance of the issues. When reporting bias, there is a complete absence of processes allowing complainants to remain anonymous or use pseudonyms, both of which are important mechanisms to avoid reprisals, redress power imbalance and shift the focus from credentials to particulars. This lack of anonymity has a particular impact on disabled individuals, who face issues around disclosure of their disability (See 5.3).

2.6.3 Researchers who are not awarded funding are left without the resources needed to progress their work, publish papers and present data to other scientists. It may also result in their institutions increasing their administrative and teaching loads via workload allocation models. If minoritised researchers are discriminated against in the grant application or peer review process, preventing them from being awarded funding, this forces them into a vicious cycle, where the lack of funding impacts their research track record, and reduces their capacity to write new proposals. Hence, their probability of winning future funding is further reduced. The impact of a small amount of bias can quickly accumulate, and make a researcher’s position untenable and unsustainable. (See 5.4).

2.6.4 Other more specific issues in the funding process affect particular minoritised groups, and the reader is referred to our publications on the specific issues facing women[21], LGBTQ+ people[22], disabled people[23], and scientists who belong to minoritised ethnic groups[24]. All of these issues affect the sustainability of careers in academic research to the detriment of retention of minoritised scientists in the STEM workforce.

3. The implications of these groups being underrepresented in STEM roles in academia and industry

3.1 Before we address the question of the implications of lack of diversity in STEM roles in academia, we note first that - irrespective of those implications - this is a question of fairness. STEM academics from minoritised groups face barriers which are not experienced by their non-minoritised counterparts when applying for public money to perform scientific research. This discrimination and inequality of opportunity is simply wrong and is liable to face legal challenges under the Equality Act 2010.

3.2 In addition to these basic questions of fairness, the lack of diversity in STEM has negative implications for innovation. A recent study used automated text-based analysis to look for markers of innovation across 1.2 million doctoral theses published in the USA between 1977 and 2015[25]. The study found that minoritised scientists are more innovative than their majority counterparts, but that they receive less reward for their new ideas and inventions. This suggests that increasing the amount of funding available to minoritised scientists can support the government’s objective to “Build back better through innovation”[26].

3.3 As well as the broad implications of lack of diversity for innovation, there are also specific implications about what is studied and how. For example, in the COVID-19 pandemic, Black people have died at a much higher rate than white people, an issue which was addressed by UKRI in a specific call for research proposals. Despite many Black investigators applying to lead projects, none received funding[27]. The projects which were funded focussed on genetic and biological approaches, overlooking the systemic and social factors which the Black researchers had proposed to investigate. This tunnel vision could have serious consequences for the health and wellbeing of the UK’s Black communities, and prevent the development of a robust approach to addressing health disparities in the future.

4. What has been done to address underrepresentation of particular groups in STEM roles

4.1 Although there are a number of initiatives which address underrepresentation of minoritised people in STEM, few of these focus on funding, particularly in the UK. Whilst initiatives such as the Athena Swan scheme and the Race Equality Charter are having some impact on the academic sector, they fail to acknowledge or address the controlling role research funding has in STEM careers.

4.2 Small changes to research funding policy can be transformative. When the Irish Research Council introduced gender-anonymised assessment of funding applications in 2014, the proportion of early career fellowships awarded to women rose from 35% in 2013 to 44% in 2014 and to 57% in 2017[28]. (See 5.5)

4.3 However, research funder policy changes require careful design. The US National Institutes of Health (NIH) have a long-standing and ongoing record of disproportionately low funding rates for Black applicants. Approaches adopted to reduce these disparities have included increased bias-awareness training, but there is growing evidence that this is ineffective[29]. In contrast, actions aimed at improving early career researcher success rates, including adjusted criteria and additional guidance to reviewers, have reduced funding disparities between early and other career stage applications at NIH29. The UK's EPSRC has stated that its evidence suggests that its own similar adjustments to New Investigator Awards[30] have led to "an increase in the ambition and quality of some projects, as well as a broadening of the range of early career researchers who now qualify for support." While any impact of this increased breadth is not evidenced by the data released by EPSRC thus far, the approach of using targeted changes to funding processes to modify criteria, reviewer guidance, and approaches to budget management point to strategies that could be successfully applied to address other disparities in research funding in the UK.

4.4 The National Institute for Health Research (NIHR) introduced incentives to encourage institutions to achieve Athena Swan accreditation for dismantling barriers to women's progress. As a result, the proportion of NIHR funding going to women increased, and there was a related increase in the number of women in mid-level leadership positions[31]. The government has recently required that NIHR cease this practice, despite its proven effectiveness[32]. (See 5.6).

4.5 Although UKRI have instituted some policies to encourage increased diversity[33], at the time of writing they have yet to publish a full Equality and Diversity strategy, although a consultation version has very recently been released (on 13th January 2021)[34]. On November 11 2020, Professor Dame Ottoline Leyser, Chief Executive at UKRI, told the Science and Technology Committee that the UKRI Equality Strategy would be published “alongside the people and culture strategy that BEIS is developing”[35]. This commitment was reiterated in April 2021[36]. The BEIS Research and development (R&D) people and culture strategy was published in July 2021[37], but no UKRI Equality Strategy accompanied it.

4.6 The BEIS strategy document is the most relevant positive action the current government has taken to address diversity in STEM, but cherry-picks data in order to conceal diversity issues. For example it highlights that in 2018/19, 3% of UKRI’s awards to co-investigators went to Black researchers, equalling the proportion of Black people in the UK population, but the document underplays the very low representation of Black people as Principal Investigators - i.e. the people who actually lead the research. In both 2017/18 and 2018/19 there were only 10 grants awarded to Black PIs across the whole of UKRI's remit, whereas around 2000 awards were made to white PIs in each year (see also 2.2). “Racism” is not even mentioned in the BEIS strategy, let alone addressed with appropriate action plans, nor are the words “bias”, “prejudice” or “discrimination”[38]. Overall, there is little evidence that either BEIS or UKRI are prioritising equality and diversity, despite the stark inequities in the sector.

5. What could and should be done by the UK Government, UK Research and Innovation, other funding bodies, industry and academia to address the issues identified.

5.1 Overcoming institutional gatekeeping

Given the role that institutions play in deciding who should have the opportunity to apply for funding (see 2.6.1), both through formal internal selection and other more informal gatekeeping processes, it is vital that funders discourage practices which reduce the number of applicants from minoritised groups. We recommend that the government should insist that research funders who disburse public money both maintain transparently available records of the diversity of applicants from individual institutions and also provide financial incentives to encourage institutions to increase the proportion of applicants from groups which are under-represented in the applicant pool.

5.2 Recording diversity data

A complete analysis of the impact of funder policy on the diversity of the STEM workforce requires the recording of data on the demographics of not only grant applicants and awardees, but also researchers employed using grant funding and funded postgraduate students. Data recorded by UKRI and other research funders has historically been patchy, and there has been significant campaigning activity focused on improving its accuracy and granularity. Although improvements are being seen, there are still holes in the data; in particular, no data related to LGBTQ+ identities are collected, and gender is treated as binary. Moreover, the impact of intersectional identities (e.g. being Black and disabled) are not analysed. A barrier to intersectional analysis can be the very low numbers of people who embody these intersections. In this case, qualitative data -- including narrative accounts of lived experience -- is vital to understanding the relevant issues. We recommend that the Government should not finalise financial settlements for research funders who disburse public money, unless those funders have specific and appropriately budgeted plans in place both to collect quantitative data across the full range of protected characteristics under the Equality Act 2010 and to conduct qualitative data collection and analysis to better understand equity issues and to help address intersectional marginalisation. All these data should be available to the research community.

5.3 Acting on diversity data

The data we describe in 5.2 will be useless, unless they are used to drive action to address the disparities in funding rates for minoritised scientists. The public sector equality duty, under the Equality Act 2010, requires funders to act to eliminate unlawful discrimination and advance equality of opportunity. The government should take steps to ensure that research funders fulfil this duty. We recommend that an independent body, or auditor, should be set up to scrutinise the diversity data issued by research funders. This body should have the power to demand improvements to data collection and analysis (see 5.2), but should also be able to require funders to produce and implement action plans to address any observed underfunding of minoritised groups, and to monitor the effectiveness of those action plans. Such an auditor could also take responsibility for independently investigating complaints about reviewer bias and other diversity issues (see 2.6.2). Overall, this would provide much-needed scrutiny of whether funders are meeting their public sector equality duty.

5.4 Universal basic research income

The cumulative impact of bias on the careers of minoritised scientists is large, and although we could recommend any number of small policy changes, more ambitious change is needed to achieve meaningful improvements to the diversity of the STEM workforce in academia. One such concept is the creation of a new income stream, which routinely provides all STEM researchers who are currently eligible to apply for grants with sufficient funding to maintain a small, continuous research programme, with grant funding being used to top up this basic income. This would help prevent the vicious cycle (see 2.6.3), in which lack of grant income prevents grant application success. However, such a scheme would not promote equity unless it included financial incentives to encourage Universities to employ minoritised researchers and recruit minoritised PhD students. We hence recommend that the government institutes a Universal Basic Research Income[39] funding stream, which provides a minimum income equivalent to one fully-funded PhD studentship to each eligible academic researcher each year, where the demographics of University STEM department academic staff and PhD students are representative of the UK population as a whole. Where the demographics are non-representative, a financial penalty should be applied for each “missing” researcher from a marginalised group, and Universities should be required to make up the shortfall from other budgets.

5.5 The Advanced Research and Invention Agency (ARIA)

The government has a stated ambition to make the UK a “science superpower”, and one of its key strategies to achieve this is the creation of ARIA, a new funding agency. Given the risks to genuine innovation arising from the current limited diversity of the STEM workforce and STEM funding awardees (see 3.2 - 3.3), and the evidence that careful design of funding policies and processes can increase diversity (see 4.2 - 4.3), the creation of ARIA presents a key opportunity to use funding policy to increase diversity in STEM and thus encourage innovation. We recommend that before plans for ARIA are finalised, a thorough review should be conducted of international best practice in funding policy as a tool for increasing diversity in STEM, and that such best practice should be implemented at ARIA from the outset.

5.6 Ensuring Institutional Accountability

In the absence of more ambitious policy changes such as those outlined above (5.1 - 5.4), the Athena Swan and Race Equality charters offer options for ensuring institutional accountability (see 4.4). Whilst both these charter marks have attracted criticism - including for the same lack of attention to qualitative data and the lived experience of minoritised academics which characterises the analyses undertaken by funders - both are being adapted and evolved to improve their relevance and effectiveness[40]. However, at present, there are no charter marks focused on disability equity or the rights of LGBTQ+ scientists. We recommend that the government should leave open to research funders the option of requiring Athena Swan and Race Equality Charter accreditation to enforce institutional accountability for improving equity. We recommend that the government introduce charter marks specific to the higher education context for Disability Equity and LGBTQ+ inclusion, which could be built on the current “Disability Confident Employer Scheme”[41] and the Stonewall “Diversity Champions”[42] programme respectively.

Overall, these measures are designed to incentivise funders and universities to greatly increase their efforts to ensure that research funding is distributed on an equitable basis, providing increased research and employment opportunities to under-represented groups. Solving inequity in research funding is key to improving diversity in the academic STEM sector.

The following members of TIGERS were involved in the preparation of this document and agreed to its submission. All are writing in a personal capacity and not offering the views of their employers.

Dr Idris Ajia, University of Southampton

Dr Claudia Antolini, Science communicator

Dr Jessica Boland, University of Manchester

Dr Sean Collins, University of Leeds

Dr Linda Cremonesi, Queen Mary University of London

Francesca Firth, Cambridge University

Dr Esther Garcia-Tuñón, University of Liverpool

Dr Izzy Jayasinghe, The University of Sheffield

Julie Jebsen, University of Manchester

Dr Neha Mehta, Queen’s University Belfast

Dr Caroline Müllenbroich, University of Glasgow

Dr Katie Nicoll Baines, University of Edinburgh

Prof Rachel Oliver, Cambridge University

Prof João Quinta da Fonseca, University of Manchester

Dr Madina Wane, graduate of Imperial College London

(January 2022)