Written Evidence Submitted by Imperial College London

(DIV0072)

Author List

Dr Sarah Essilfie-Quaye, Dr Julia Makinde, Professor Francisco Veloso, Dr John Tregoning, Dr Pei F Lai

Executive Summary

We acknowledge the submission of complementary evidence by our colleagues from Imperial College London which outlines the series of measures Imperial College London has taken to address challenges faced by underrepresented groups.

In this report, we focus specifically on:

A) The reasons why ethnic minorities are underrepresented in STEM roles in academia

B) The implications of ethnic minority groups being underrepresented

C) 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

 

This report mainly focuses on people from the Black ethnic group, however with increasing seniority there is increased effect/disadvantage on all ethnic minority groups. We identified three key reasons why ethnic minorities are underrepresented in STEM: The Broken Pipeline, Staying Power, and Systemic inequalities.

 

The implications of these groups being underrepresented are: loss of talent, suboptimal scientific output, suboptimal clinical training and delivery of clinical services and systemic bias in the development of technology.

 

We recommend the following actions to address the issues identified:

 

 

 

 

About the authors of this document

The authors are academics and researchers at various career stages at Imperial College London. For the most part they are involved in cross-institutional efforts aimed at improving diversity and equal access in academic research and industry. Author biographies are provided as an appendix within this document.

Key Evidence and Recommendations

In evaluating the evidence presented in this we recommend that the committee considers the following

 

 

A) Reasons for underrepresentation in STEM roles in academia

i) The Broken pipeline
The pipeline that ensures representation in STEM roles in academia and industry relies on adequate representation in early training stages (e.g., the achievement of STEM A-levels, access to undergraduate, postgraduate, and early training schemes in industry).

  1. There are disparities in access to research intensive universities by minoritized groups. Ethnic minority pupils were found to constitute 26% of all undergraduate admissions. However, they continue to be underrepresented in Russell Group universities, with Black students found to be particularly underrepresented, 3.3% Russell Group vs 8.1% New Universities (1,2). This highlights the need to define the demographics that fall under the umbrella term “ethnic minorities” as this often prevents adequate assessment of the factors that are at play within a group that is far from being homogenous.
  2. Outreach efforts aimed at improving access for minoritized groups to University training programmes are often disproportionately borne by staff belonging to these groups with little remuneration or recognition that leads to promotion for the effort (3). The Stuart Hall Foundation in partnership with the Centre on the Dynamics of Ethnicity prepared an excellent review (in 2021) of 13 previous reports dating between 1981 and 2017 (spanning 37 years) looking into inequalities that affect ethnic minorities in the UK. A total of 589 recommendations have so far been made according to their report, many of which have yet to be taken up.
  3. Once in the university training system, evidence suggests that minoritized groups are less likely than white students to receive a 2:1 or first-class degrees (1). This likely has an impact on the progression of these graduates to academic and industry careers. This point was also highlighted in the 2019 Report which focused on “The Broken Pipeline: Barriers to Black PhD Students Accessing Research Council Funding” (4). The system also fails to capture the full context of the scientific curriculum as it relates to the experiences and contributions of minoritized groups. This problem is compounded by the dearth of Black Lecturers and potential role models, causing trainees to feel like they do not belong within the system.
  4. The 2021 Hamilton Commission (5) found that minoritized groups had disproportionately poor access to industry careers due to the industry focus on recruitment of the very “best” from select or “elite” universities. Given that access to these universities (e.g., Russell Group) is disproportionately reduced for minoritized groups, this selection bias further enhances the attainment gap.
  5. The Hamilton 2021 Commission noted additional barriers (logistical and economic) that prevent minoritized groups from getting into the apprenticeship schemes that serve the pool of technical experts that support the motorsport industry. This is likely repeated in other engineering apprenticeship schemes.
  6. A poor understanding, knowledge or exposure to industry or academic careers or role models amongst ethnic minorities has also been highlighted by all the above as factors that contribute to the broken pipeline.

ii) Staying Power
A failure to feed the pipeline at its earliest stages ensures exclusion of minoritized groups from the onset. Where they exist in the system, the evidence suggests that it is an uphill task to:

 

  1. Less than 1% of UK Professors are Black. That is 155 out of 23,000 (6). Evidence documented on the lived experiences of 20 (out of 25) Black Female Professors (7) identified the following barriers to career progression:
  1. The “snowy white peaks” of the NHS: a survey of discrimination in governance and leadership and the potential impact on patient care in London and England (8) showed that there remains a very significant gap between the composition of Trust Boards and national NHS bodies, and the rest of the workforce and the local population. Research evidence suggests this may well adversely impact on the provision of services, denying the NHS the potential contribution that a diverse leadership could make. 
  2. Black engineers in the UK motorsport industry interviewed by the Hamilton Commission expressed a sense of frustration in progression through to leadership roles, highlighting microaggressions and racist banter that they were compelled to “put up with”. These experiences and a general lack of commitment from industry leadership likely contributes an exodus of the little talent that exists in the industry.

 

iii) Systemic and structural inequalities
The following systemic factors have been identified:

  1. Several reports (1) cited above collectively highlight the stark and historic absence of data to enable meaningful discourse, and the willingness to objectively analyse and address the systemic drivers of underrepresentation and exclusion.
  2. Funding disparities. In 2020 for the first time (9), the UKRI published its analysis of the ethnicity data for UK Research and Innovation funding applicants and awardees. Updated data reported in 2021 (9) showed that whilst the proportion of fellowship awardees from ethnic minorities had increased, fellows specifically from the Black ethnic group still constituted less than 1% of awardees. They also account for less than 1% of Principle Investigators, which is well below the HESA estimate of academic and labour market share.
  3. UKRI admits to not having full ethnicity information for UKRI funded PhD studentships. Where data exists, ethnic minorities account for 13% of candidates, with the majority being Asian.
  4. Lack of transparency in funding processes. The Head of the UKRI responded in 2021 to an open call to address systemic and structural inequalities that prevented Black Scientists from accessing funds linked to addressing disparities in COVID-19 outcomes in the UK BAME community, listing 6 areas in which UKRI will seek to address the concerns of Black academics on its funding processes and outcomes.
  5. Lack of institutional support. Some funding applications require a percentage/matched contribution or costs towards large equipment for applicants to secure research funding. Institutions are more likely to fund senior academics, of which minoritized or Black researchers make a small percentage. This disparity constitutes a significant attainment barrier.
  6. Structural/environmental drivers. The Royal Society reports about the widening attainment gap as ethnic minority group STEM qualifiers (from university) progress into employment, stating that a higher percentage of ethnic minority group STEM leavers were found to be unemployed 6 months after graduation compared to White STEM leavers (10).
  7. The Royal Society also reports that minoritized groups are more likely than their peers to be on fixed term (temporary) contracts. They were also more likely to earn less than their peers at the same job level (11).

B) The implications of these groups being underrepresented in STEM roles in academia and industry

  1. Loss of talent: The exclusion of excellent candidates very early in the pipeline results in the loss of talent and critical perspectives in all sectors and industries. When they get through the early stages, individuals who should normally transition into more senior roles within these sectors opt to leave academia and industry.
  2. Suboptimal scientific output: Reduced diversity of researchers contributes to the fact that much of the (clinical) research conducted in the UK does not involve key groups. The 2020 NIHR INCLUDE report concludes that it is impossible to generalise the results from studies excluding key groups or to inform public policy that arises from the same. It can also mean that prescribed medical and public health interventions have different or contradictory effects in the excluded groups (12). Participation must cut across all stakeholders (academics, scientists, clinicians), to ensure that the science is as robust as it can be (13,14).
  3. Suboptimal clinical training and delivery of clinical services: Where the curriculum is not inclusive, as evidenced by the need for Mind the Gap handbook, the ability of clinical trainees to adequately deliver services to the UK population is adversely impacted (15).
  4. Systemic bias in the development of technology: Inadequate representation in technological research has led to systemic (AI /coding) bias that makes minorities targets for the wrong reasons. Recent reports (16) from the US suggests that facial recognition algorithms used by the FBI had an error rate of 15% in 2017. A 2018 MIT report also found that software aimed at facial recognition misidentified dark skinned women up to 35% compared to 1% for light-skinned men. These errors have been attributed in part to the training dataset used in the development of these technologies (16). Another example is the use of pulse oximeters which are widely used in clinical practice without much consideration for the fact that they have been shown to overestimate arterial oxygen saturation during hypoxia in dark-skinned individuals (17). This likely exacerbated the poorer clinical outcomes observed in BAME patients infected with SARS-CoV-2 (18).

C) 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

  1. Data is key: Government policy where possible should mandate or normalise the recording and monitoring of diversity data across all areas highlighted in this submission. Analysing the statistics will enable multilevel reflection and help define targets with measurable actions to be taken, consequences for inaction or missed targets.
  2. Representation Matters: Ensure representation across panels and decision committees in academia and industry. Ensure the experiences of those affected are considered, making sure their contribution is valued, because the pressure on researchers from ethnic minority groups to participate in diversity activities likely impacts other career metrics (3). If this is adequately addressed, it will serve to correct biases in the system.
  3. Fix the processes: This will enable clarity with respect to obtaining funding, training, recruitment, and promotion within academia, with support structures built alongside this to help individuals navigate successfully. Bureaucracy associated with, for example, funding applications, can disadvantage underrepresented staff (19) if they don't have the right network of people to support them by giving them access to advice and useful connections. Updating policies to reduce needless bureaucratic barriers would be helpful to minoritized groups.
  4. Race Equality Charter: Mandate REC[2] participation at all Higher Education Institutions in the United Kingdom. After the introduction by NIHR of Athena SWAN incentives, the number of women theme-leads as well as the proportion of women in senior director positions increased (20). This is a great example of positive policy-based actions being effective.
  5. Government/Industry partnerships: Policy and government initiatives could aim to partner with industry representatives who have shown a keen interest in addressing these inequalities to develop targeted schemes that aim to address the problems with the pipeline, focusing initially on the early/entry stages and progressively ensuring that staying power is enhanced.
  6. Targeted hiring policies: Whilst the subject of targeted hiring may be somewhat controversial, we suggest that it may be a way to correct systemic bias and ensure that ALL qualified candidates are adequately considered. This is especially the case when given evidence that inequality undermines the selection for excellence (21). The most recent hire (and arguably one of the most qualified) on the UK show “Countdown” was recently described on social media as a token Black hire in the wake of the George Floyd murder and the BLM protests. We suggest instead that following the events of 2020, the show executives were now more open to explore the possibility that a Black woman may be perfectly qualified to co-host the TV show.
  7. Repopulating the scientific curriculum: Encourage/fund initiatives to accelerate the repopulation of the content and contextual relevance of the UK STEM curriculum to ensure that it communicates scientific concepts that are relevant to all.

Reference list

  1. 2021 SHF Race Report 2021. The Stuart Hall Foundation
  2. 2020 ETHNICITY, RACE AND INEQUALITY IN THE UK State of the Nation
  3. Gewin V. The time tax put on scientists of colour. Nature. 2020;583(7816):479-481.
  4. 2019 The Broken Pipeline – Barriers to Black PhD Students Accessing Research Council Funding.
  5. 2021 Accelerating Change: improving the Representation of Black People in UK Motorsport.
  6. 2021 Higher Education Staff Statistics: UK 2019/20
  7. 2019 Staying Power. The career experiences and strategies of UK Black female professors.
  8. 2014 The “snowy white peaks” of the NHS: a survey of discrimination in governance and leadership and the potential impact on patient care in London and England Kline, Roger.
  9. a)2021 Detailed analysis of UKRI funding applicants and awardees ethnicity, financial years 2015-16 to 2019-20.

b)2020 Detailed analysis of UKRI funding applicants and awardees ethnicity, financial years 2015-16 to 2018-19.

  1. 2020 and 2021 Baselines for Improving STEM Participation: Ethnicity STEM data for students and academic staff in higher education 2007/08 to 2018/19 Will Joice and Andy Tetlow 15 October 2020 Conducted on behalf of the Royal Society
  2. 2018 Ethnic minority academics earn less than white colleagues BBC Report Croxford 
  3. 2020 Improving inclusion of under-served groups in clinical research: Guidance from INCLUDE project; 07 August 2020
  4. 2020Diversity wins: How inclusion matters McKinsey Report 2020
  5. 2016 Why Diverse Teams Are Smarter Rock
  6. 2020 Mind the Gap: a handbook of clinical signs on black and brown skin
  7. 2020 Big brother watch briefing on facial recognition surveillance
  8. Bickler PE, Feiner JR, Severinghaus JW. Effects of skin pigmentation on pulse oximeter accuracy at low saturation. Anesthesiology. 2005;102(4):715-719.
  9. Wong AI, Charpignon M, Kim H, et al. Analysis of Discrepancies Between Pulse Oximetry and Arterial Oxygen Saturation Measurements by Race and Ethnicity and Association With Organ Dysfunction and Mortality. JAMA Netw Open. 2021;4(11): e2131674. Published 2021 Nov 1.
  10. 2022 Independent Review of Research Bureaucracy, Interim Report
  11. Ovseiko PV, Taylor M, Gilligan RE, et al. Effect of Athena SWAN funding incentives on women's research leadership. BMJ. 2020;371:m3975. Published 2020 Oct 26
  12. 2015 Implicit and unconscious, the bias in us all. A Royal Society Blog.

 

List of contributors

Dr Sarah Essilfie-Quaye is the Faculty of Medicine Equality, Diversity, and Inclusion Research Fellow at Imperial College London. Her research focuses on Black early career researchers from postdoctoral through to fellowships and academic positions, and female academics progressing to senior and leadership roles. As well as investigating the inclusion of under-served groups in clinical research undertaken within the faculty.

Sarah is also a Co-Chair for Imperial As One, Imperial College London’s Race Equality Staff Network.

Dr Julia Makinde is a Senior Manager, Clinical Immunology at IAVI and Honorary Lecturer at Imperial College London. She is currently a member of the EDI committee at the Department of Infectious Disease, Faculty of Medicine, Imperial College London.

Professor Francisco Veloso is the Dean of Imperial College Business School and a professor of innovation and entrepreneurship. He has several dozen publications in leading academic journals and has won several awards for his academic work. He regularly contributes as a consultant, advisor, and non-executive director to start-ups, established firms, universities, and government around the world. 

Francisco has a PhD in Technology, Management and Policy from the Massachusetts Institute of Technology, as well as an MSc in Economics and Management of Science & Technology, and an Undergraduate Degree in Physics Engineering, both from the University of Lisbon.

Dr John Tregoning Is reader in respiratory infections in the Department of Infectious disease, where he also chairs the equality, diversity, and inclusion committee. He has been a principal investigator for 14 years.

Dr Pei F Lai is a Research Associate, who specialises in Women’s Health (Pregnancy and Parturition) within the Faculty of Medicine at Imperial College London. She is also a Personal Tutor for BSc Medical Biosciences undergraduate students, and a committee member for the Imperial As One staff network. 

 

(January 2022)

[1] BAME- Black, Asian and Minority Ethnic; PoC- People of Colour

[2] The Race Equality Charter (REC) aims to improve the representation, progression and success of minority ethnic staff and students within higher education.