Executive summary
This report presents a comprehensive description of early research career demographics and transitions in the life sciences.
Using data from the Higher Education Statistical Agency (HESA) and from Biotechnology and Biological Sciences Research Council (BBSRC) doctoral training programmes (DTPs), we examine equality, diversity and inclusion (EDI) characteristics across phases in the doctoral trajectory in life science disciplines, including:
- transition from first to postgraduate research degree, including offers of admission in BBSRC DTPs
- progression through doctoral studies
- employment destination after doctoral graduation
The main findings from this report confirm, specifically for the life sciences, issues raised by a number of commentators about inequalities in the early research career.
These issues have been highlighted for the early research career more generally by the Leading Routes report and recent open letters to UK Research and Innovation (UKRI) (Giles et al., 2020; Williams et al., 2019).
We find race and ethnic inequalities in the transition from first degree to postgraduate research degrees in life sciences subjects. These are not accounted for by academic attainment.
Black graduates in particular are less likely to be called for interview or offered a studentship, less likely to successfully complete their PhD and less likely to transition to postdoctoral academic employment after successfully completing a PhD.
These are troubling findings that reinforce the importance of action to achieve a life sciences research workforce that fairly represents UK society.
We identify other inequalities. Women are less likely to progress to a life sciences PhD than men, as are graduates from post-1992 universities and those who have non-graduate parents.
Women, Black or Black British graduates and those gaining their PhD from a post-1992 university were all less likely to proceed to a postdoctoral position immediately after their PhD. Additionally, our data shows a gender wage gap for life sciences doctoral graduates in their first roles post-PhD.
Our analysis points to the role university status hierarchies and regional disparities play in shaping opportunities.
Life sciences doctoral students are concentrated in a relatively small set of universities, which also tend to have higher numbers of first-degree students from advantaged backgrounds and lower numbers from minority ethnic groups.
We found relatively low levels of movement across types of university between undergraduate and doctoral level, suggesting that there are structured pathways that work against graduates from the post-1992 institutions.
There is a risk that inequalities are ‘baked in’ to institutional hierarchies.
Regional mobility is also reduced compared to first-degree level. Most regions are net exporters of graduates into life sciences PhDs, except for London and the East of England, which benefit from ‘brain gain’.
Given the importance of life sciences research and development in the UK economy, this further underlines the need for redistribution of prosperity to the UK’s regions.
We found little higher education mobility between levels for life sciences degrees and, therefore, prestigious higher education institutions continued to be richer and whiter at the doctoral degree level.
Moreover, regional mobility is also reduced from the first-degree to the postgraduate research degree, although the South East of England and London have a considerable brain gain.
Taken together, we hope these findings will inform and galvanise efforts by BBSRC, DTPs and collaborative training partnerships (CTPs) and the wider higher education life sciences research community to address early career inequalities.
Recommendations
This report aims first and foremost to describe patterns of progression through the early research career in the life sciences, identifying any apparent inequalities.
Nevertheless, the issues we draw out require attention, and we offer the following recommendations for addressing them.
1. Create an EDI partnership framework to focus and coordinate BBSRC action
We suggest this should contain at least three elements.
First, it should connect to work at UKRI level on EDI issues in research, including UKRI’s external advisory group for equality, diversity and inclusion.
Second, it should involve engagement with expert groups and the life sciences community. As we note, some of our findings and suggestions repeat those already raised recently in reports and open letters written by academics from minority groups (for example, Giles et al., 2020; Williams et al., 2019).
Including these academics in the discussion is the first step to understanding persistent race and ethnic inequalities in early research careers and creating the appropriate initiatives to tackle the problem.
Third, to be most effective a mechanism is needed to coordinate action across BBSRC, DTPs and potentially other partners (other funders, learned societies, etc.).
2. Establish doctoral studentship schemes in the life sciences for minority ethnic groups, as proposed by the ‘Broken Pipeline Report’ (Williams et al., 2019)
There are stark race inequalities in access to postgraduate research programmes and funding. First-degree graduates from minority groups are less likely to immediately transition to postgraduate research degrees when compared to White students.
Moreover, they are less likely to be invited to an interview and offered a BBSRC studentship. While longer-term actions are necessary, there is a pressing need for positive action to avoid continued inequality.
Careful coordination is required to ensure that schemes are appropriately designed to fit equalities legislation and to maximise impact within and between research organisations.
3. Introduce targeted postdoctoral schemes for minority ethnic groups.
Individuals who identified as Black or Black British are less likely to have a postdoctoral contract three months after graduation.
To address inequalities in the life sciences early research career, it will not be sufficient to address access to the PhD, not least because of the time this will take to have an effect. Positive action on postdoctoral positions is also needed.
4. Address structural barriers to opportunity
Many of the graduates from underrepresented groups complete their undergraduate study in post-1992 universities.
Those graduates must be better connected with PhD opportunities, through outreach from DTPs, through expanding the reach of DTPs to include more post-1992 universities, or both.
Additionally, BBSRC, UKRI and the Westminster and devolved governments must consider the regional distribution of life science funding and opportunities.
5. Invest in developing understanding of early research career inequalities
There is a need for better and more joined-up data about doctoral study and doctoral students in the life sciences, and about the postdoctoral career inside and outside academia.
Beyond descriptive data, there is a need to better understand the processes by which inequalities arise. And there is a need to evaluate interventions to understand what works and what does not.
Introduction
Making progress on equality, diversity and inclusion is a founding aim of UKRI’s organisational strategy.
The UKRI Chief Executive, Professor Dame Ottoline Leyser, has called for an “inclusive knowledge economy” where “everyone has the opportunity to participate, and from which everyone benefits”.
It has been further highlighted through the Black Lives Matter debate in relation to underrepresentation of students from certain ethnic groups, especially Black students among UKRI studentship holders (Williams et al., 2019).
In 2020, a widely circulated open letter raised concerns about student recruitment in UKRI DTPs and CTPs.
These concerns in the research community reflect broader public debate about inequalities and prejudice on the basis of race, ethnicity and gender, and about the opportunities for those from socio-economically and geographically marginalised groups, to be socially mobile.
The current UK government has declared that it is pursuing a ‘levelling up’ agenda to distribute opportunities more evenly around the country.
These debates and agenda are highly relevant to UK life sciences. The life sciences are a prominent feature of the UK’s economic and research landscape, and the renewed emphasis on research as a key element of post-EU Exit national strategy point to the continuing importance of the life sciences research workforce.
In addition to arguments of fairness and equality, it is vital that life sciences careers are open to those with the talent and potential to enter them, from any background.
Furthermore, a life sciences research workforce that reflects the make-up of the UK will ensure that the future agenda for life sciences research incorporates diverse views and values to better understand and address the scientific challenges of the coming decades.
While there is a growing understanding of the broad patterns of entry to postgraduate research (as the first major step in a research career), such analyses are very much in the aggregate, even where they focus on science, technology, engineering, and mathematics (STEM) disciplines.
Retention of scientific researchers in academic research is low, as the Royal Society has demonstrated (The Royal Society, 2010), but understanding of early career transitions remains shaky.
Better data is needed, but insufficient use has been made of some of the extensive datasets already available.
We do know that the structure of early careers in research varies considerably across broad fields. Such variations are seen in the timing and component parts of the early career, such as whether qualifying degrees are typically taken immediately following preceding qualifications or after graduates have spent time in the labour market.
They are seen in the institutional concentration or dispersion of doctoral and postdoctoral research and funding, and in the extent to which early transitions dictate much later outcomes.
Understanding this structure and the disciplinary culture in which it is located is a prerequisite to formulating when, where and how to intervene to address inequalities.
We must begin by understanding
A basis for action on equality, diversity and inclusion in the UK life sciences must therefore begin with an understanding of the field.
In this report, we present a wide-ranging description of early career demographics and transitions in the life sciences, using the best available extensive datasets.
While the report covers the life sciences in general, we foreground BBSRC disciplines and students as a key element of the field in the UK, particularly in relation to doctoral training.
Furthermore, BBSRC has declared its intention to ensure fair and equitable access to its funding opportunities in the early career and beyond.
Doctoral training partnerships and centres funded by BBSRC have also initiated and developed conversations and action intended to widen doctoral participation, with some beacons of good practice emerging.
There is thus both a top-down and bottom-up impetus from within BBSRC’s research community for a thoroughgoing equality, diversity and inclusion agenda.
In this report, we aim to contribute to this impetus by describing patterns of early-career participation and progression in the life sciences.
This descriptive analysis covers four principal phases.
First, we outline transition from a first degree into a higher degree by research in the life sciences as a graduate outcome.
We are able to show how this varies according to a range of factors including gender, ethnicity, socioeconomic class, prior attainment and location (institutional and geographical).
Second, we present data on rates of application, offer and enrolment within BBSRC DTPs to highlight how these vary, sometimes in troubling ways, across applicant characteristics.
Third, we investigate progression through a research degree in the life sciences across EDI characteristics, where a more positive picture arises.
However, in our final set of analyses, which follows doctoral graduates into the labour market, we see some concerning patterns around entry into academic jobs and EDI measures.
Taken together, these different datasets allow us to piece together the landscape of the early research career in the life sciences.
The report is structured as follows. We begin by describing and explaining our datasets, noting their strengths and limitations. We then proceed with our analyses of the different phases of the life sciences early research career. We conclude the report with a broader discussion of the findings.
Data and methods of analysis
In this section of the report, we describe the datasets on which our analyses are based and the variables they contain.
These datasets are, to the best of our knowledge, the most comprehensive and extensive datasets available to investigate EDI in the life sciences early research career.
We focus in the report on UK-domiciled students (although EU-domiciled students are included in some analyses). Our focus is therefore on those students who, at the time of enrolling into higher education, were ordinarily resident in the UK.
Many people come to the UK to study life sciences or to join the life sciences research workforce.
We do not wish to imply, in our focus on UK-domiciled individuals, that EDI issues do not affect international staff and students. Rather, our focus on UK staff and students provides a consistent basis for comparison across the phases.
Moreover, there is considerable heterogeneity among international staff and students that makes it very difficult to establish a base population to compare with (for instance when looking at socioeconomic class or ethnicity).
Datasets
We use three datasets sourced from the Higher Education Statistical Agency (HESA) to examine inequalities in access and within doctoral programmes in the life sciences field of studies.
HESA is the statutory body for the collection of data about publicly funded and regulated higher education institutions in the UK.
It has been collecting significant amounts of census and survey data on UK students and graduates for over 25 years.
HESA has robust procedures for assuring data quality and coverage. The data analysed here was supplied under licence through HESA’s partnership with Jisc, the UK digital, data and technology agency focused on tertiary education, research and innovation. We acknowledge the advice, support and guidance from Jisc colleagues.
The first dataset, which will be referred to henceforth as the transition dataset, comprises all UK- domiciled first-degree graduates between the academic years 2012 to 2013 and 2016 to 2017.
The transition dataset contains information on individuals’:
- characteristics and socioeconomic background
- first-degree subject, attainment and higher education institution
- first destination after graduation
For individuals who enter postgraduate education after graduation, the dataset has information on the type of qualification and higher education institution attended at that level.
The second dataset, referred to as the progression dataset, comprises all UK and EU-domiciled doctoral entrants in 2011 to 2012 and 2012 to 2013 and tracks them either to successful completion, withdrawal or failure, or continued enrolment by 2018 to 2019.
The third dataset, referred to as the destination dataset, comprises all UK-domiciled doctoral graduates between 2012 to 2013 and 2016 to 2017.
The destination dataset includes detailed information on their first destination (employment) three months after graduation and, for a small subset of 2012 to 2013 graduates, their destination three years after graduation. Finally, aggregate application data has been supplied to us directly by BBSRC.
This dataset collates data on doctoral applications, applicants invited to interview and offers made between the years 2016 to 2017 and 2019 to 2020 across BBSRC DTPs and CTPs.
Data is available broken down by certain EDI characteristics (disability, sexual orientation, gender and age). It is not disaggregated into data for each DTP or CTP.
Variables
Within the HESA datasets, we use the Joint Academic Coding System (JACS) to identify individuals in the life sciences undergraduate and postgraduate degrees in the three datasets.
The JACS consists of a code that identifies the subject group, the principal subject and discipline.
The principal subjects in life sciences funded by BBSRC are:
- agriculture
- agricultural sciences
- animal science
- biology
- botany
- broadly based programmes within biological sciences
- broadly based programmes within veterinary sciences agriculture and related subjects
- clinical veterinary medicine and dentistry
- forestry and arboriculture
- genetics
- microbiology
- molecular biology biophysics and biochemistry
- others in biological sciences
- others in veterinary sciences agriculture and related subjects
- pre-clinical veterinary medicine
- sport and exercise science
We also use the JACS to identify medical and engineering and physical sciences subjects, which are funded, respectively, by the Medical Research Council (MRC) and Engineering and Physical Sciences Research Council (EPSRC), to compare characteristics of doctoral programmes and graduates’ outcomes.
To understand higher education institutional inequalities, we have used an established categorisation of the prestige of the UK higher education institutions (Wakeling & Mateos-Gonzalez, 2021). The groupings are as follows:
- Oxbridge: comprises Oxford and Cambridge universities
- other golden triangle: includes prestigious higher education institutions based in London (Imperial College, King’s College, University College London and the London School of Economics. These institutions, plus Oxford and Cambridge, are often referred to as the ‘golden triangle’ based on their geographical locations)
- other Russell Group: comprises the higher education institutions that are part of the group, except Oxbridge and the golden triangle institutions
- non-Russell Group pre-1992: includes higher education institutions that were created before the Further and Higher Education Act 1992 and are not part of the Russell Group
- post-1992: institutions that were granted the university title after the 1992 act
We do not wish to imply any endorsement of institutional prestige hierarchies. However, it is important to recognise that such hierarchies are ‘out there’ and can be observed in different ways in policy and in data (Boliver, 2015; Wakeling & Savage, 2015).
Analysis
All data presented in this report complies with the HESA standard rounding methodology.
The rounding methodology consists of rounding all numbers to the nearest multiple of 5 and suppressing percentages based on less than 22.5 individuals and averages based on 7 or fewer individuals. The purpose of the methodology is to avoid identifying individuals in the datasets.
We use average marginal effects (AMEs) to report the regression models presented in the report. AMEs are an intuitive way to interpret regression coefficients for lay readers since they are standardised to allow ready comparison.
In the linear regression models, the AME represents the average difference in the outcome for each predictor’s categories and its reference group.
In the logistic and multinomial logistic models, the AMEs represent the difference in the predicted probability between each predictor’s categories and its reference group.
Negative AMEs indicate a smaller average or lower probability, while positive AMEs indicate a higher average or probability when compared to the reference groups.
Limitations
Few datasets are perfect, and ours are no exception. As noted in the next section, we are only able to analyse certain background characteristics for those graduates who move immediately from a first degree into a research degree.
Where there is a break between the two levels, including where a graduate proceeds into a research degree via a taught masters, the data linkage is broken.
Our dataset does not contain the very latest available academic years, and where we are investigating outcomes longitudinally, it necessarily refers to research students who began their degrees some time ago.
As in all secondary data analysis, we must rely on variables and definitions we did not create, and which might not be ideal for our current purposes.
Unfortunately, we are not able to identify, in the transition dataset, which graduates entered a research degree funded by a research council studentship.
For the application data, we have only aggregate statistics to work with, which limits the depth of analysis possible.
We know that there is some non-response to the destinations survey we use in the transitions and outcomes datasets (although not substantial) and that data coverage on some variables is better than others.
Finally, we acknowledge that the disciplinary ‘borders’ of the life sciences are porous. There is movement both into and out of life sciences academic units between first-degree and research degree study, often from or into ‘adjacent’ disciplines.
Since we need to use standardised disciplinary categories with our data, this unavoidably loses some nuance.
For example, a single-honours chemistry graduate entering a biochemistry PhD would, in our dataset, count as changing discipline, whereas a single-honours biology graduate would not.
Similarly, a PhD student in a school of life sciences might mainly be engaged with mathematics or computing science research, rather than life sciences per se. Nevertheless, we believe we have used the best data readily available to us for our analyses.
Transition of first-degree graduates into postgraduate research study
To examine the effect of socioeconomic background on the first destination for UK-domiciled first-degree graduates, we use the transition dataset.
We use the coarsened exact matching (cem) method (Iacus et al., 2012) to match every individual who transitioned to a postgraduate research degree in the life sciences (treatment group) with individuals with the same undergraduate academic background characteristics (control group).
The cem method reduces imbalance in the covariates between treatment and control group, reducing model dependence and increasing power and efficiency (smaller confidence intervals and standard errors).
The matching procedure allows us to temporarily coarsen the covariates (in our case, academic background) and apply the exact matching for all individuals in the treatment group.
Individuals are then sorted into strata, and individuals in a stratum without treated units are excluded from the analysis.
In the analysis, a weight is used to account for individuals in the treatment group who were matched with more than one individual. The cem method was performed using the ‘cem’ Stata command (Blackwell et al., 2009).
To measure academic background, we use variables for University and Colleges Admissions Service (UCAS) tariff, level of qualification and principal subject of the first degree.
Therefore, the control group comprises first-degree graduates with the same academic credentials as those who entered a postgraduate degree in the life sciences subjects. Appendix A compares first-degree graduates in the life sciences subjects who pursue a clinical and an academic route immediately after graduation against EDI characteristics.
Data only tracks immediate transition of graduates
It is important to highlight that the data used in this section only allows us to track immediate transition after the first degree.
Therefore, we are only comparing differences in the characteristics and socioeconomic background for first-degree graduates who immediately transitioned to a postgraduate research degree in the life sciences with their academically equivalent peers who did not.
We do not capture individuals with a non-consecutive transition to a postgraduate research degree (for example, individuals who first pursue a master’s degree or enter the labour market).
Nevertheless, according to HESA data, a large number of students in the life sciences postgraduate research degree transition directly from undergraduate (see table below).
| Academic year | Transition dataset | First year postgraduate research |
|---|---|---|
| 2012/13 | 960 | 1,045 |
| 2013/14 | 1,050 | 1,130 |
| 2014/15 | 905 | 1,190 |
| 2015/16 | 1,110 | 1,140 |
| 2016/17 | 1,190 | 1,290 |
| Total | 5,215 | 5,795 |
Characteristics of graduates transitioning into research study
Sex of graduates transitioning
The majority of first-degree graduates immediately transitioning to a postgraduate degree in life sciences are female (respectively, 60.2% of those not in postgraduate research and 52.2% of those in postgraduate research).
Nevertheless, the odds of immediately transitioning to a postgraduate research degree are higher for male first-degree graduates (1 in 234 for males and 1 in 325 for females), showing gender inequalities in access to a postgraduate degree in the life sciences.
Thus, while males are (just) in the minority of those transitioning to a research degree, they are progressing at a rate quite a bit higher than we would expect.
This overrepresentation is not unique to life sciences (Leslie et al., 2015), but it may be somewhat hidden compared to the more obvious underrepresentation of women in other STEM fields.
Race and ethnicity of graduates transitioning
Research on race and ethnic inequalities has shown that first-degree graduates from minority groups are more likely to pursue a postgraduate taught degree immediately after graduation as a mechanism to avoid unemployment or underemployment (for example, Lessard-Phillips et al., 2018; Wakeling, 2009).
Nevertheless, little is known about the impact of race on the transition to a postgraduate research degree.
Individuals who declared being Black or Black British, Asian or Asian British (Indian, Pakistani and Bangladeshi) and Other Asian backgrounds are less likely to transition to a research degree than other ethnicities.
Based on the number of first-degree graduates academically qualified, we would expect three times the number of Black or Black British and about twice the number of Asian or Asian British and Other Asian backgrounds.
The comparison also shows that UK first-degree graduates of Chinese and unknown ethnicities are more likely to transition to a life sciences research degree than those of other ethnicities.
Self-reporting of disability among graduates transitioning
A recent study shows that PhD students in the UK are more likely to report some level of mental health problems when compared to the general public (Hazell et al., 2021).
Over 70% of the doctoral research students in the sample reported having depression and anxiety symptoms during the PhD.
Nevertheless, at the time of transition from first degree to postgraduate research degree, only 12.4% of the students who transitioned to a postgraduate research degree in the life sciences reported having any disability (including mental disability).
Moreover, there is not any clear indication that students who reported having a disability (mental, physical and unknown disability) have different probabilities of pursuing a postgraduate research degree when compared to students who did not report having any disabilities.
Socioeconomic classification of graduates transitioning
Figures for the number and the difference between the expected and actual number of individuals who transitioned to a life sciences postgraduate research degree by socioeconomic background use the National Statistics Socioeconomic Classification (NS-SEC).
The NS-SEC is based on the occupation of the highest earner in the graduate’s household. For graduates under 21 years old, the NS-SEC of the parent or guardian’s household is recorded, and, for students 21 years old and over, their own household NS-SEC is recorded. Information on the NS-SEC is compulsory for UK-domiciled students (HESA, 2021a).
First-degree graduates who have a socioeconomic background of higher managerial, administrative and professional occupation are more likely to transition to a postgraduate research degree in life sciences when considering the expected number.
Moreover, first-degree graduates from intermediate and routine and manual occupations are less likely to transition to a postgraduate research degree immediately after the first degree.
Parental education of graduates transitioning
A higher proportion of first-degree graduates who immediately transitioned to a postgraduate research degree in a life sciences discipline have a parent with a higher education degree (53.1%) when compared to first-degree graduates with similar academic qualifications (44.4%).
First-degree graduates with a parent with a higher education degree might have more knowledge of different academic trajectories and their advantages and risks, therefore, they might have invested in the required skills and dispositions to pursue a postgraduate degree.
The positive association between socioeconomic background and transition to a postgraduate degree in the life sciences is consistent with the finding on inequalities in access to postgraduate education in the UK and European countries (for example, Wakeling and Hampden-Thompson, 2013; Wakeling and Laurison, 2017).
Type of first higher education institution for graduates transitioning
The table below presents the type of higher education institution attended for the first degree by whether individuals transitioned to a postgraduate research degree in life sciences subjects.
As was shown in the recent study by the Institute of Fiscal Studies (Van Der Erve et al., 2021), there are large inequalities in access to selective higher education institutions at the undergraduate level in the UK, and higher education institution has a significant impact on labour market outcomes and, consequently, social mobility.
Therefore, it is important to understand the institutional trajectory of postgraduate research students.
| HEI type | Number not in postgraduate research | % | Number in postgraduate research | % | Total | % |
|---|---|---|---|---|---|---|
| Oxbridge | 21,790 | 1.48 | 430 | 8.28 | 22,220 | 1.51 |
| Other golden triangle | 29,670 | 2.02 | 235 | 4.54 | 29,905 | 2.03 |
| Other Russell group | 332,835 | 22.67 | 2,335 | 44.71 | 335,165 | 22.75 |
| Non-Russell group pre-1992 | 340,960 | 23.23 | 1245 | 23.88 | 342,205 | 23.23 |
| Non-Russell group post-1992 | 742,645 | 50.59 | 970 | 18.58 | 743,615 | 50.48 |
| Total | 1,467,895 | 100 | 5,220 | 100 | 1,473,110 | 100 |
The majority of first-degree graduates who transitioned to a postgraduate degree studied in a Russell Group institution (57.5%), while 50.6% of the first-degree graduates who did not transition to a postgraduate research degree studied in post-1992 institutions.
This table indicates that the type of higher education institution plays an important role in transitioning to a postgraduate research degree in the life sciences subjects.
Key points
Among those who are similarly qualified, women, and Black and Asian first-degree graduates transition to a postgraduate research degree in the life sciences at a lower rate than expected. Men and White graduates have a higher-than-expected rate of transition.
The association between socioeconomic background and transition to a life sciences postgraduate research degree is similar to that found in previous research. Taking into account academic background, those with parents in professional or managerial employment and of graduate status are more likely to proceed to a research degree.
While graduates of post-1992 universities make up the majority of first-degree graduates in the life sciences, the majority of first-degree graduates who transitioned to a life sciences postgraduate research degree studied in a Russell Group university for their undergraduate degree.
Type of higher education institution for postgraduate research degrees in life science subjects
The effect of UK higher education institutional stratification on graduates’ outcomes is not exclusive to undergraduate degrees, but also has an impact on the postgraduate level (Pásztor & Wakeling, 2018; Wakeling & Savage, 2015).
This section investigates the association between the type of higher education institutions of postgraduate research degree and individuals’ characteristics and socioeconomic background for those who transitioned to a postgraduate degree after the undergraduate degree.
Of the 5,220 first-degree graduates who immediately transitioned to a postgraduate research degree in the life sciences, 62.2% attended a Russell Group, 18.5% a pre-1992 and 13.9% a post-1992 higher education institution.
The distribution of the type of higher education institution for UK-domiciled first-degree graduates who immediately transitioned to a postgraduate research programme (transition dataset) is similar to the distribution for UK-domiciled doctoral entrants in 2011 to 2012 and 2012 to 2013 (progression dataset).
Nevertheless, the latter has a higher proportion of individuals in non-Russell Group pre-1992 higher education institutions.
Institutional mobility between first and postgraduate research degrees
There is a strong association between the type of higher education institution of the first degree and the type of higher education institution of the postgraduate research degree.
Individuals who immediately transitioned to a doctoral programme in the life sciences subjects tend to study in the same or similar higher education institution to the first-degree.
Even though there is a small drift ‘upwards’ (individuals who attended a non-Russell Group pre-1992 and post-1992 institution in their first degree entering other Russell Group institutions for their postgraduate research degree and individuals who attended other Russell Group institutions entering Oxbridge and other ‘golden triangle’ institutions), in general, there is very little long-range mobility (up or down).
In other words, there is not much institutional mobility in total between undergraduate and postgraduate research degrees.
Characteristics and socioeconomic backgrounds of postgraduate students by type of higher education institution
Race inequalities among individuals who transitioned to a postgraduate research degree are significant in all types of higher education institutions.
Nevertheless, other golden triangle higher education institutions have a higher proportion of Black, Asian and minority ethnic (BAME) backgrounds when compared to other types of institutions, which might be explained by their location.
According to the UK Census 2011, individuals from BAME backgrounds are more likely to live in London than any other region in the country: 58.4% for Black, 35.9% for Asian, 33.1% for Mixed and 49.9% for other backgrounds (UK Census, 2011). The other higher education institutions in London have similar proportions of BAME postgraduate research students.
Regarding individuals’ socioeconomic backgrounds, students who transitioned to a postgraduate degree in the life sciences in Oxbridge universities tend to be from more privileged socioeconomic backgrounds when compared to other types of higher education institutions. They have a higher proportion of parents in higher managerial and professional occupations (40%) and at least one parent with a higher education degree (66%).
Regional mobility of transitioning graduates
There is substantial mobility between the region of the domicile and region of the first degree, with a large influx of first-degree students in life sciences to Scotland, Yorkshire and the Humber, North East of England, and Wales.
Nevertheless, while there is not as much regional mobility between first and postgraduate degrees, there is a ‘brain gain’ for London and the East of England and ‘brain drain’ in Yorkshire & the Humber, South East of England and Wales.
Key points
The majority of first-degree students who transitioned to a postgraduate degree immediately after graduation enrolled for their doctoral degree in a Russell Group university.
There is a high association between the type of higher education institution of the first and postgraduate degree.
Inequalities at the undergraduate level are maintained at the postgraduate research level. Postgraduate research students in Oxbridge universities are from more advantaged socioeconomic backgrounds when compared to other types of higher education institutions.
Student regional mobility is higher from domicile to first degree than from first degree to a postgraduate research degree. Nevertheless, there is a brain gain in the postgraduate research in the East of England and London.
Doctoral applicants
To fully understand entry to the early career, it is not sufficient only to look at those who ended up enrolling.
We also need to understand who sought to enrol and, for this, we used anonymous doctoral applicant data between 2017 to 2018 and 2019 to 2020 provided by BBSRC.
We examine the profile of applicants to the PhD programmes through DTPs and their relative rates of success on being invited to an interview and receiving an offer of a place to study (see Appendix B for the list of the lead organisations and full partners in 2021).
After analysing the data on doctoral applicants, we did not find any inequalities regarding sex, sexual orientation, or disability when comparing the number of applications, interviews and offers. Nevertheless, as shown by ‘The Broken Pipeline’ report (Williams et al., 2019), there are substantial race and ethnic inequalities in access to BBSRC-funded DTPs.
First, we compared the number of applicants to BBSRC’s DTPs with the reference population of qualified first-degree graduates (from the transition dataset) to assess the representativeness of the applicant pool.
Even though there is a higher percentage of White graduates academically qualified, the distribution shows that minority groups are proportionately represented in the application pool. However, the relatively high proportion of unknown ethnicity applicants makes interpretation more challenging.
Next, we compared the probability of being invited for an interview and having a successful application for each ethnicity group.
Individuals who declared being White and of ‘Other’ ethnicity have a higher probability of a successful application (respectively, 16.3% and 18.9%) when compared to minority groups.
In contrast, for individuals who declared being Black (N=313) or Asian (N=959), respectively, 20.1% and 26.5% were invited for an interview and 8.0% and 8.1% had a successful application for BBSRC’s DTPs programme.
Therefore, Black and Asian applicants have half of the probability of receiving an award from these programmes when compared to White applicants.
There is a clear trend when looking at applications by age. The older the applicant, the less likely they will be invited for an interview and receive funding for their doctoral programme.
Key points
The proportion of applicants from ethnic minority groups to BBSRC DTPs is broadly representative of the population of first-degree graduates.
Black and Asian applicants are less likely to be invited for an interview or have a successful application than White applicants.
Older applicants are less likely to be funded by BBSRC.
Doctoral programme
A marginally higher proportion of UK-domiciled individuals among life sciences doctoral graduates between 2012 to 2013 and 2016 to 2017 (destination dataset) were in the Oxbridge, pre-1992 and post-1992 higher education institutions when compared to medical and engineering and physical sciences doctoral graduates.
Regarding the major source of fees for the doctoral programme, 28.3% of the graduates in life sciences did not receive any award or did not have any financial backing.
Also, 28.3% of the life sciences graduates received a research council or British Academy scholarship, compared to 15% of medical and 32% of engineering and physical sciences graduates. The category ‘research council & British Academy’ is that used by HESA. The British Academy previously provided research studentship funding in the arts and humanities, but this role has subsequently been transferred to AHRC.
We also examined the association between life sciences graduates’ sex, ethnicity, type of higher education institution for the postgraduate research degree, and major sources of fees using a logistic regression model.
We look at the average marginal effects for the categories “research council & British Academy” and “no award or financial backing”.
While there is no gender difference in the probability of not being self-sponsored, female graduates were less likely to be research council funded after controlling for ethnicity and type of higher education institution.
Black and Black British graduates were more likely to be self-sponsored and less likely to have a research council award when compared to White graduates in the life sciences, confirming the concerns raised in the open letter to UKRI (Giles et al., 2020) and the finding from the ‘Broken Pipeline’ report (Williams et al., 2019).
Moreover, there are striking institutional inequalities on graduates’ major source of fees for the postgraduate research degree.
Graduates from Oxford and Cambridge universities were more likely to receive research council funding than any other type of higher education institution.
As has been shown in this report, doctoral students from the Oxbridge institutions and UKRI councils were also more likely to be individuals from privileged socioeconomic backgrounds (parents in managerial occupations and with a higher education degree).
Key points
Black or Black British postgraduate research students are less likely to receive research council funding in the life sciences when compared to White postgraduate research students, after controlling for gender and type of higher education institution.
Postgraduate research students from Oxbridge are more likely to receive research council studentship funding compared to other types of higher education institutions.
Progression of doctoral students in the life sciences
Using the dataset on the UK and EU-domiciled doctoral cohorts in the life sciences in 2011 to 2012 and 2012 to 2013 for all UK higher education institutions, we track doctoral students through and out of their programme and examine the difference in trajectories by socioeconomic background, gender, ethnicity, the major source of fees and type of higher education institution.
About 18% of the doctoral students in the life sciences did not finish their studies. We did not find any gender inequalities regarding dropout among life sciences doctoral students.
While the difference in the percentage of students who finished and who dropped out by whether the students are from White, BAME and Unknown backgrounds is not substantial, the disaggregated variable for ethnicity indicates that students who identify as Black and British are more likely to drop out compared to other minority groups.
In line with the HESA rounding methodology, we cannot present any percentages based on fewer than 21 individuals.
Students who self-reported having mental disabilities in the first year of the PhD are more likely to drop out (23.5%) when compared to students who did not report any disabilities.
Due to the small number of cases, we cannot present the data for students who reported having a physical disability and unknown disability type.
It is important to highlight that students reported their disabilities in the first year of the PhD programme. Therefore, it does not account for students who received a diagnosis or developed any disability during their studies.
Funding sources and dropout rates
There is a link between funding source and completion rates when looking at the percentage of dropouts by major source of funding and type of higher education institution. Research council-funded students are most likely to complete and self-funded students are least likely to complete.
Similarly, students at Russell Group universities, especially Oxbridge, have a higher completion rate than those at other institutions.
As seen in the previous section, there is a strong correlation between the two measures. Students from Oxbridge are more likely to be research council funded, while students from post-1992 universities are more likely to be self funded.
Differences in time taken to complete research
In this section, we also analyse differences in the number of years taken by students who successfully finished their PhD in the life science subjects.
The majority of students finished their doctorate studies in four years (61.8%). 17.4% finished their studies in less than 4 years and 20.8% finished in 5 years or more.
Our analysis does not show any socioeconomic and gender difference in the number of years to finish the PhD programme.
Unfortunately, because of the small number of cases, we could not examine disaggregated ethnic differences. Although a higher percentage of BAME students finished their PhD in 5 years or more (25% compared to 20% of the White students), we cannot point to substantial differences in the number of years to finish the PhD by ethnicity.
Even though the type of disability might have a different impact on academic trajectories, when we considered the years to finish the PhD programme, we saw few differences between students who self-reported having mental, physical and other disabilities.
Therefore, to avoid suppressing information due to the small number of cases, we look at the number of years to finish the PhD in the life sciences for the aggregated variable for disability.
Students who reported having mental, physical or other types of disabilities are more likely to finish their PhD studies in five years when compared to students who did not report any disabilities.
When comparing students’ major source of fees, we found some substantial differences in the number of years to finish the PhD studies.
For students funded by the research councils, 71.4% finished their postgraduate research degree in four years. For students who did not have any award or financial backing, 39.1% finished in four years and 37.7% finished in five years or more.
The type of higher education institution is also associated with the number of years to finish the PhD in the life sciences programmes.
While the majority of students in Oxbridge, ‘golden triangle’, Russell Group and pre-1992 institutions finished their PhD in four years, the majority of students in the post-1992 higher education institutions finished their PhD in five years or more (44.1%), which might be explained by their higher percentage of part-time students.
We are unable to investigate how full-time or part-time study modes impacted on completion times as over 70% of research students in our dataset qualified from the status of ‘writing up’.
Key points
The majority of doctoral students in the life sciences finish their PhD in four years.
Around one-fifth of entrants to life sciences PhDs leave their studies without gaining a PhD. While numbers are small, Black and Black British students are less likely to finish.
The majority of students who were funded by the research councils finish their PhD in four years or less.
Students in the post-1992 sector and self-funded students are more likely to finish their PhD in five years or more.
Destinations of doctoral graduates in the life sciences
To examine outcomes from life sciences doctorates for UK-domiciled graduates, we used the destination dataset, which comprises UK-domiciled doctoral graduates between 2012 to 2013 and 2016 to 2017.
The data collected by HESA is the most comprehensive longitudinal data on PhD graduates available in the UK. Nevertheless, it has several limitations (Hancock, 2021; Hancock et al., 2019).
Due to the high amount of missing information, we cannot examine the effect of socioeconomic background (parental occupation and occupation) on life sciences PhD graduates’ destinations.
Moreover, the data includes information on destination three years after graduation for only a small number of cases. We have therefore, decided to briefly present these results separately, in Appendix C.
We explored the intersectionality between gender and ethnicity on the destination of the doctoral graduates in the life sciences. Nevertheless, because of the small number of cases and the report space limitation, we decided not to present them in the report.
To summarise, we found some indication that males from minority groups are less likely to be in a postdoctoral contract after graduation or employed in the education sector, while females from minority groups are less likely to be in managerial or professional occupations three months after graduation.
We examine the frequency distribution for the variables of postdoctoral contract, academic job, occupation and salary by students self-reported disability for doctoral graduates.
We found some differences between the type of disability and whether the PhD graduates were in a postdoctoral position and whether they were in a managerial or professional occupation three months after graduation.
We cannot present the frequency tables on graduates’ destination and type of disability because of the small number of cases (only 6.4% of the doctoral graduates in the life sciences reported having a disability).
Consequently, the coefficients were not statistically significant when we included the disaggregated variable for students’ disabilities on the regression models.
Employment information for life sciences graduates
About 80% of the graduates in the life sciences doctoral programmes were in professional or managerial occupations three months after doctoral completion.
Almost 55.5% of the graduates in life sciences took up employment in the education sector three months after graduation and about 20% were in professional, scientific and technical activities.
Moreover, 55.6% were on a fixed-term contract and 39% were on a permanent or open-ended contract after graduation.
These results confirm trends found in Organisation for Economic Co-operation and Development (OECD) countries, which show an increase in the number of doctoral degree holders in temporary contracts after graduation and a higher engagement of doctoral graduates from natural sciences and engineering in research after graduation when compared to doctoral graduates from social sciences (Auriol et al., 2013).
Academic jobs
Those few studies that examine the destination of doctoral graduates focus on understanding the transition to academic employment (Hauss et al., 2015; Hottenrott & Lawson, 2017; van de Schoot et al., 2012), with little known about transition to industry (Germain-Alamartine et al., 2021; Kyvik & Olsen, 2012).
Almost all studies show that departmental and organisation factors have a significant impact on whether doctoral graduates enter the academic labour market, but less focus is placed on graduates’ characteristics and socioeconomic background.
Moreover, although there is an established debate on the representation of women in STEM in academia, much less is known about how ethnic minorities’ representation in academia might affect the number of ethnic minority postgraduate research students, nor those students’ intentions to pursue an academic job after graduation.
So far, we have shown that there are race and ethnic inequalities in access to a postgraduate research degree and studentship funding in the life sciences.
In this section, we want to contribute to the discussion on minority group representation in academia by examining how individuals’ characteristics might affect their likelihood to enter an academic job after graduation.
To understand whether individuals’ gender, ethnicity, and type of higher education institution of the doctoral programme affect their likelihood of pursuing an academic career after graduation and its employment basis, we use two different variables for academic jobs.
The first is an indicator variable for individuals with a postdoctoral contract three months after graduation. This information is derived from graduates’ answers regarding their job titles, and it was provided in the dataset.
The second variable examines whether individuals are working in a higher education institution, including post-doctorates, three months after graduation, taking into account the type of contract.
To determine whether individuals were in an academic job, we created a new variable using occupation and economic activity.
For the former, we run a logistic regression model to estimate the probability of having a postdoctoral contract and, for the latter, we run a multinomial logistic regression model to estimate the probability of having an academic position.
On average, female individuals have a lower probability of having a postdoctoral research contract compared to male individuals after controlling for ethnicity and higher education institution type of the doctoral programme.
Moreover, individuals who declared being Black or Black British are less likely to have a postdoctoral contract when compared to White individuals, while there are no statistical differences in the probability of having a postdoctoral research contract between White and ‘Other’ ethnicities.
Last, individuals who earned their postgraduate research degree in a post-1992 higher education institution are less likely to get a postdoctoral contract.
On average, female life sciences graduates were less likely to remain in academia than male graduates.
Moreover, while Asian, Other and Mixed ethnicities show no statistically significant differences compared to White doctoral graduates, Black or Black British graduates have a higher probability of not staying in academia.
Lastly, life sciences doctoral graduates from other Russell Group, pre-1992 and post-1992 are less likely to stay in academia when compared to Oxbridge graduates.
When considering individuals who stayed in academia and have a permanent contract, ethnicity does not have a statistically significant effect.
Nevertheless, on average female doctoral graduates in life sciences are less likely to have an academic position on a permanent contract when compared to male graduates.
Moreover, graduates from pre-1992 and post-1992 are more likely to stay in academia on a permanent contract when compared to Oxbridge graduates in life sciences.
Last, there was no statistically significant effect of gender for graduates who were in an academic position with a fixed-term contract 3 months after graduation. However, Black and Black British were less likely to have an academic position on a fixed-term contract when compared to White graduates.
The same was true for post-1992 graduates when compared to Oxbridge doctoral graduates in the life sciences.
The different effects of higher education institution type on the probability of getting an academic job by contract type might be explained by differences in academic trajectories of graduates from Oxbridge and post-1992 institutions.
While graduates from post-1992 institutions might pursue more teaching-focused academic jobs immediately after the PhD, graduates from Oxbridge institutions might first pursue a postdoctoral job (where most contracts are fixed-term) and, after, apply for academic jobs in more research-intensive higher education institutions.
Nevertheless, further research should investigate differences in academic trajectories of doctoral graduates in the UK by type of higher education institutions.
Occupation and income
After controlling for ethnicity and higher education institutional type, on average women are less likely to occupy managerial or professional occupations compared to men three months after graduation.
Moreover, individuals who attend post-1992 higher education institutions are less likely to be in managerial or professional occupations when compared to individuals who attend Oxbridge universities after controlling for individuals’ characteristics.
Nevertheless, there are no ethnic and race inequalities in the likelihood of having a managerial or professional occupation for doctoral graduates in the life sciences after controlling for sex and higher education institution type.
While there are no ethnic inequalities regarding salary, on average, female graduates earn less than male graduates three months after finishing the postgraduate research degree in life sciences.
Moreover, doctoral graduates in the life sciences from other Russell Group and pre-1992 higher education institutions earn, on average, less than Oxbridge graduates.
Key points
The majority of doctoral graduates in the life sciences are employed in the education sector three months after graduation.
Female doctoral graduates are more likely to be in a non-academic position after graduation when compared to men.
Black and Black British doctoral graduates are more likely to be in a non-academic position and less likely to have a fixed-term contract three months after graduation when compared to White graduates.
Institutional differences have an effect on academic trajectories after graduation. Graduates from pre and post-1992 institutions are more likely to have an academic position with a permanent contract three months after graduation.
Female doctoral graduates in life sciences subjects are less likely to be in a managerial or professional occupation and earn, on average, less than male doctoral graduates.
Doctoral graduates from other Russell Group and pre-1992 higher education institutions earn, on average, less than doctoral graduates from Oxbridge universities.
Discussion
In this final section, we draw out the main findings from our analysis, reflect on what they tell us more broadly about inequalities in the early research career in the life sciences, and consider some potential actions to bring about change.
We wish to begin by acknowledging that many of the findings reported in our analysis confirm earlier findings in two ways.
First, researchers and activists of colour have pointed out the stark underrepresentation of Black people in particular among recipients of research council studentships, and also of research council grant funding (Adelaine et al., 2020; Giles et al., 2020; Williams et al., 2019).
We hope that UKRI and BBSRC will honour their commitment to engage with Leading Routes, the #KnowledgeisPower collective and others who have repeatedly raised these issues to call for a process of constructive change.
The Office for Students and Research England initiative to widen participation in postgraduate research for Black, Asian and minority ethnic groups is a positive step.
Finding a way for the research councils to formally associate with this initiative, or to set up their own will help to go further in addressing the striking underrepresentation of Black people among research council studentship holders and postdoctoral staff, including in the life sciences.
Second, where there is previous research to draw on, many of the patterns we report here for the life sciences are similar to patterns seen in postgraduate research participation overall.
Some of the factors and processes that influence inequalities in life sciences early research careers transcend any one discipline and hence cannot be entirely solved from within the field itself.
Indeed, some of the grander structural influences, such as institutional stratification of universities and regional economic inequalities within the UK are not easily addressed by UKRI and are instead matters for national government.
However, the nature of different disciplines, variations in how they are organised and the inequalities that affect them most all mean that the road to greater equality in early career research must pass through the disciplines.
Key messages for BBSRC and life sciences
In our view, there are five key messages for BBSRC and the life sciences community arising from our analysis.
Our analysis points to differences in the structure of opportunities available for progression into life sciences doctoral study.
When comparing those beginning a PhD with their equally qualified peers who did not, it is clear that women, students from most minority ethnic groups and from less socioeconomically advantaged backgrounds are underrepresented.
For some groups, particularly Black graduates, this underrepresentation is stark. It contrasts with the representation of graduates by ethnicity among those life sciences graduates who move on to postgraduate clinical training (for example, medicine as a second undergraduate degree).
As we show in Appendix A, Black and South Asian groups have higher levels of representation here, although the overall numbers are relatively small.
We also show that there is a strong connection between the institutional location of potential life sciences PhDs at first-degree level and the likelihood of making an immediate transition to a PhD.
The institutions with lower proportions going on to a doctorate are also the institutions with the highest proportion of underrepresented students.
Coupled with the relative lack of institutional mobility we show between first degree and PhD, this strongly suggests that DTPs need to look outside their usual institutional recruiting grounds in order to widen participation (at least to address race and ethnic and socioeconomic inequalities; gender differences do not seem to relate to institutional stratification in the same way).
As has been the case for undergraduate widening participation, there is a case for further outreach from DTPs, especially to post-1992 universities. These institutions are much less likely to host DTPs and CTPs but represent half of all life sciences first-degree graduates.
A related theme is the concentration of life sciences activity institutionally and geographically. This is a complex issue, since there are arguments in favour of clustering research rather than dispersing it, particularly in the natural sciences.
Such rationale lies behind initiatives such as the Crick Institute and Cambridge’s Silicon Fen. Nevertheless, patterns of concentration in the life sciences reflect national patterns, with an overall concentration in London and the South East (we note that BBSRC DTPs are distributed throughout the UK though).
While moving institution or sector between first degree and PhD is relatively limited, the movement that takes place points to a ‘brain drain’ from the devolved nations and other parts of England into London and the East of England.
Going beyond access, through our analysis of the progression and outcomes datasets we identify what looks like ‘U-shaped’ inequality across the life sciences early career.
Access to doctoral study is unequal. Once a doctoral programme is entered, progression through the programme shows few substantial inequalities in completion between different groups, with the notable exception that the relatively small number of Black PhD students are less likely to complete the programme.
However, we see inequalities re-appear in doctoral destinations. Women and Black doctoral graduates are less likely to enter academic jobs after their PhD, according to our data.
From the data we have, we cannot tell why there might be such differences. Nevertheless, there is a clear need for action, especially on ethnicity.
If there are lower than expected numbers of Black graduates entering life sciences PhDs, and Black doctoral graduates are less likely to enter an academic job, then as well as acting on access there is a need to do something to reverse the postdoctoral situation.
There are different options for action
There are different options for action to address the issues raised in this report. We would expect that action across several areas and by a range of different groups is needed.
Some actions may be beneficial across the different inequalities identified. Better targeting of first-degree graduates in the life sciences in post-1992 universities, for instance, would help address race and ethnic and socioeconomic inequalities as well as potentially ameliorating regional differences.
Actions may have a different balance of risk and reward. ‘Outreach’ by DTPs is fairly straightforward but might not make significant inroads into the inequalities identified.
Conversely, adopting quotas for doctoral or postdoctoral funding for underrepresented groups could be challenging, but has been shown to lead to dramatic change in other countries.
In Brazil, for example, the Quota Law, implemented in 2016, assigned 50% of the places in public funded higher education institutions to students from Black, mixed-race, or indigenous backgrounds who studied in public funded secondary schools.
A recent report on one of the most prestigious higher education institutions in the country showed that the number of undergraduate students from minority background almost doubled after the policy was implemented (Paula et al., 2021).
While the UK and Brazil have very different demographics, history and higher education systems, the example shows that radical actions can lead to change. Many BBSRC DTPs are already taking forward their own actions in doctoral access.
We know, through the wider project of which this report forms a part, that there is commitment and determination to address equality, diversity and inclusion in the life sciences among DTPs and CTPs.
Other funders are introducing schemes too, such as the Wellcome Sanger Institute (Cambridge) and its postdoctoral scheme for Black heritage scholars.
Positive action schemes like this providing dedicated awards for students from underrepresented groups will be an important means of supporting early career researchers at PhD and postdoctoral levels.
Careful coordination of schemes is needed across and beyond the life sciences research community to facilitate maximum impact for such schemes.
Experience of previous initiatives is that they tend to be somewhat atomised and unconnected. This makes systematic evaluation of impact much more difficult to achieve.
Putting in place a mechanism to connect institutions and initiatives, perhaps through BBSRC, could help in this regard.
Examining selection practices and criteria is another potential means of addressing underrepresentation.
Research on equity in graduate admissions in the US shows that the adoption of particular notions of ‘academic pedigree’ as a sign of merit can have the effect of excluding underrepresented groups from positions at the shortlisting stage (Posselt, 2016).
Examples might include the use of research experience as an indicator of ability (rather than of prior opportunity) and of first-degree university as a proxy for ability.
In case study research, US institutions that successfully widened access to their graduate school shared a common approach: revision of their selection criteria and practices (Posselt, 2016, 2020).
We need better-quality data
Finally, our experience of conducting the research and analysis for this report has reconfirmed the need for more and better-quality data about the research career life course.
Compared to key comparator countries such as Germany and the US, we lack rich longitudinal data about the postdoctoral research career (Hancock et al., 2019).
Tracking PhD graduates is limited to their ‘first destination’ six to 18 months after completion, with only a small number tracked beyond that.
Most of the data we have is factual, with little systematic data on motivations, career progression and change, and so on.
Similarly, data about postgraduate students, including PhD students, is less extensive than for undergraduates, especially in relation to background characteristics and prior academic experiences.
Connecting individual students’ records across different qualifications as they leave and re-enter higher education is possible, but only at some expense to researchers.
Statutory bodies such as UKRI have a role to play here in prioritising the collection and collation of data, especially for EDI characteristics.
We hope that the evidence presented in this report will contribute to both discussion and action to improve representation and inclusion in the life sciences early career.
References
Adelaine, A., Kalinga, C., Asani, F., Ngozika Agbakoba, R., Smith, N., Adisa, O., Francois, J., King-Okoye, M., Williams, P., & Zelzer, R. (2020, August). Knowledge Is Power—An Open Letter To UKRI.
Auriol, L., Misu, M., & Freeman. (2013). Careers of Doctorate Holders: Analysis of Labour Market and Mobility Indicators (OECD Science, Technology and Industry Working Papers No. 2013/04; OECD Science, Technology and Industry Working Papers, Vol. 2013/04). https://doi.org/10.1787/5k43nxgs289w-en
Blackwell, M., Iacus, S., King, G., & Porro, G. (2009). Cem: Coarsened Exact Matching in Stata. The Stata Journal: Promoting Communications on Statistics and Stata, 9(4), 524–546. https://doi.org/10.1177/1536867X0900900402
Boliver, V. (2015). Are there distinctive clusters of higher and lower status universities in the UK? Oxford Review of Education, 41(5), 608–627. https://doi.org/10.1080/03054985.2015.1082905
Garrud, P. (2011). Who applies and who gets admitted to UK graduate entry medicine? – An analysis of UK admission statistics. BMC Medical Education, 11(1), 71. https://doi.org/10.1186/1472-6920-11-71
Germain-Alamartine, E., Ahoba-Sam, R., Moghadam-Saman, S., & Evers, G. (2021). Doctoral graduates’ transition to industry: Networks as a mechanism? Cases from Norway, Sweden and the UK. Studies in Higher Education, 46(12), 2680–2695. https://doi.org/10.1080/03075079.2020.1754783
Giles, S., Bond, A., Britton, B., Ford, H., Greene, S., Grieve, S., Jackson, C., Lengger, S., Littler, K., & Pancost, R. (2020). Open Letter Concerning CDT and DTP Student Recruitment Across the UK [Open Letter]. https://docs.google.com/document/d/1EInAKFl7px2DxYv-sAZxVOyTjkm-AE29CzFFnjxw6hg/edit
Hancock, S. (2021). What is known about doctoral employment? Reflections from a UK study and directions for future research. Journal of Higher Education Policy and Management, 43(5), 520–536. https://doi.org/10.1080/1360080X.2020.1870027
Hancock, S., Wakeling, P., & Chubb, J. (2019). 21st Century PhDs: Why we need better methods of tracking doctoral access, experiences and outcomes (p. 1040511 Bytes). Research on Research Institute. https://doi.org/10.6084/M9.FIGSHARE.9917813
Hauss, K., Kaulisch, M., & Tesch, J. (2015). Against all odds: Determinants of doctoral candidates’ intention to enter academia in Germany. International Journal for Researcher Development, 6(2), 122–143. https://doi.org/10.1108/IJRD-07-2014-0015
Hazell, C. M., Niven, J. E., Chapman, L., Roberts, P. E., Cartwright-Hatton, S., Valeix, S., & Berry, C. (2021). Nationwide assessment of the mental health of UK Doctoral Researchers. Humanities and Social Sciences Communications, 8(1), 1–9. https://doi.org/10.1057/s41599-021-00983-8
HESA. (2021a). National Statistics—Socio-Economic Classification. https://www.hesa.ac.uk/support/data-intelligence/national-statistics-socio-economic-classification-200203-200809-200910
HESA. (2021b). Rounding and suppression to anonymise statistics | HESA. https://www.hesa.ac.uk/about/regulation/data-protection/rounding-and-suppression-anonymise-statistics
Hottenrott, H., & Lawson, C. (2017). Flying the nest: How the home department shapes researchers’ career paths. Studies in Higher Education, 42(6), 1091–1109. https://doi.org/10.1080/03075079.2015.1076782
Iacus, S. M., King, G., & Porro, G. (2012). Causal Inference without Balance Checking: Coarsened Exact Matching. Political Analysis, 20(1), 1–24. https://doi.org/10.1093/pan/mpr013
Kyvik, S., & Olsen, T. B. (2012). The relevance of doctoral training in different labour markets. Journal of Education and Work, 25(2), 205–224. https://doi.org/10.1080/13639080.2010.538376
Leslie, S.-J., Cimpian, A., Meyer, M., & Freeland, E. (2015). Expectations of brilliance underlie gender distributions across academic disciplines. Science, 347(6219), 262–265. https://doi.org/10.1126/science.1261375
Lessard-Phillips, L., Boliver, V., Pampaka, M., & Swain, D. (2018). Exploring ethnic differences in the post-university destinations of Russell Group graduates. Ethnicities, 18(4), 496–517. https://doi.org/10.1177/1468796818777543
Pásztor, A., & Wakeling, P. (2018). All PhDs are equal but … Institutional and social stratification in access to the doctorate. British Journal of Sociology of Education, 39(7), 982–997. https://doi.org/10.1080/01425692.2018.1434407
Paula, G. B. de, Nonato, B. F., & Nogueira, C. M. M. (2021, November). Equalização do acesso à UFMG após uma década de ações afirmativas [Equility in access to UFMG after a decade of affirmative action]. Nexo. https://pp.nexojornal.com.br/opiniao/2021/Equaliza%C3%A7%C3%A3o-do-acesso-%C3%A0-UFMG-ap%C3%B3s-uma-d%C3%A9cada-de-a%C3%A7%C3%B5es-afirmativas
Posselt, J. R. (2016). Inside graduate admissions: Merit, diversity, and faculty gatekeeping. Harvard University Press.
Posselt, J. R. (2020). Equity in science: Representation, culture, and the dynamics of change in graduate education. Stanford University Press.
The Royal Society. (2010). The Scientific Century: Securing our future prosperity (Summary Report) [RS Policy document 02/10]. Royal Society. https://royalsociety.org/-/media/Royal_Society_Content/policy/publications/2010/4294970126.pdf
van de Schoot, R., Yerkes, M., & Sonneveld, H. (2012). The Employment Status of Doctoral Recipients: An Exploratory Study in the Netherlands. International Journal of Doctoral Studies, 7, 331–348. https://doi.org/10.28945/1718
Van Der Erve, L., Drayton, E., & Britton, J. (2021). Which university degrees are best for intergenerational mobility? The IFS. https://doi.org/10.1920/re.ifs.2021.0202
Wakeling, P. (2009). Are Ethnic Minorities Underrepresented in UK Postgraduate Study? Higher Education Quarterly, 63(1), 86–111. https://doi.org/10.1111/j.1468-2273.2008.00413.x
Wakeling, P., & Mateos-Gonzalez, J. L. (2021). Inequalities in the Higher Degree? Postgraduates, prices and participation (p. 68). The Sutton Trust.
Wakeling, P., & Savage, M. (2015). Entry to elite positions and the stratification of higher education in Britain. The Sociological Review, 63(2), 290–320. https://doi.org/10.1111/1467-954X.12284
Williams, P., Bath, S., Arday, J., & Lewis, C. (2019). The Broken Pipeline: Barriers to Black PhD Students Accessing Research Council Funding. Leading Routes. https://leadingroutes.org/mdocs-posts/the-broken-pipeline-barriers-to-black-students-accessing-research-council-funding
Appendices
Appendix A. First-degree graduates who followed a clinical route immediately after graduation
Following the research performed by Garrud (2011), this appendix examines the differences by EDI characteristics between first-degree graduates in life sciences by whether they pursued:
- a clinical route
- a bachelor’s in medicine and dentistry or subjects allied to medicine immediately after graduation
- an academic route, students who pursued a postgraduate research degree in any field of study immediately after
It is important to highlight that the postgraduate research group in this comparison is different from the group analysed throughout the report (postgraduate research students in the life science doctoral programmes).
Of the 126,925 first-degree graduates in life science between 2012 to 2013 and 2016 to 2017, about 1.0% (N = 1,410) pursued a clinical route and about 4.6% (N = 5,805) pursued an academic route.
As shown in the main report, the majority of graduates in life sciences degrees are female. Nevertheless, we found a higher percentage of female first-degree graduates among those who pursued a clinical route (58.8%) than an academic route (53.5%).
First-degree graduates from minority ethnic groups are more likely to pursue a clinical than an academic route, corroborating the findings from Garrud (2011).
Nonetheless, we found that, in addition to a higher percentage of Black or Black British first-degree graduates (5.6%), the clinical route seems to be more attractive to first-degree graduates from Asian or Asian British backgrounds (16.1% against 3.7% in the academic route).
We also found a negligible percentage of students from Chinese backgrounds following a clinical route.
Regarding students’ disability and socioeconomic background, we found little difference between first-degree graduates pursuing a clinical or academic route.
Appendix B: Lead organisations and full partners of BBSRC DTPs
Lead organisation is listed first.
University College London, Birkbeck University, Kings College London, LSHTM, QMUL, Royal Holloway of London, Royal Veterinary College, University of Greenwich
University of Oxford, Oxford Brookes University, The Pirbright Institute, Diamond Light Source, ISIS Neutron and Muon Source, Research Complex Harwell, STFC Central Laser Facility, The Rosalind Franklin Institute
The University of Warwick, University of Birmingham, University of Leicester, Aston University, Harper Adams University
University of Liverpool, Newcastle University, Durham University
The University of Manchester
University of Bristol, University of Bath, Cardiff University, University of Exeter, Rothamsted Research
John Innes Centre, University of East Anglia, Quadram Institute Bioscience, Earlham Institute, The Sainsbury Laboratory
The University of Edinburgh, University of Aberdeen, University of Dundee, University of St Andrews, University of Stirling, Scotland’s Rural College, IBioIC, James Hutton Institute, Moredun Research Institute
University of Reading, Aberystwyth University, Brunel University London, Cranfield University, Queens University Belfast, University of Surrey
University of Southampton, University of Kent, University of Sussex, University of Portsmouth, NIAB-EMR
University of Nottingham, Nottingham Trent University, National Biofilm Innovation Centre
University of Leeds, University of York, The University of Sheffield
Appendix C: Destination three years after graduation
The dataset includes information on destination three years after graduation for only 330 postgraduate research graduates in the life sciences.
Moreover, the dataset includes a small number of individuals who were declared to be Black and Black British and Other ethnic backgrounds. Therefore, we cannot examine race inequalities for destination three years after graduation.
About 90% of the postgraduate research degree graduates in the life sciences were in a professional or managers, directors, and senior official occupation.
The model did not find any statistically significant effect of individuals’ characteristics and type of higher education institution of the doctoral programme on occupation.
Most graduates continue in the education sector three years after graduation (50%), followed by professional, scientific, and technical activities (26%).
Moreover, the majority of graduates are on a permanent or open-ended contract after graduation (62%).
There was no statistically significant difference in the probability of having an academic job three years after graduation by individuals’ characteristics and type of higher education institution of the postgraduate research degree.