UKRI diversity data for funding applicants and awardees 2020 to 21 update

Summary

UKRI publishes annual diversity data on the funding profile of applicants and awardees to aid transparency in our funding and enable high-level monitoring of trends. This report provides data for the 2020 to 2021 financial year and adds to our previous publications describing diversity data and detailed ethnicity analysis for the 2019 to 2020 financial year.

The report primarily presents a high-level annual overview of our diversity data at UKRI level.

Interpreting our diversity data is complex, however, as it is based upon thousands of applications to hundreds of funding opportunities administered by 7 different research councils. Annual changes in the types of opportunity launched, or external factors such as the COVID-19 pandemic, impact upon our data and the interpretation of trends over time.

Where possible, data for individual research councils is provided in accompanying data tables. Figures from analysis at council level, funding stream level or funding opportunity level are also discussed where they can inform our understanding of UKRI-level trends. Building this more detailed understanding is vital to support priority 6.2 (make UKRI an efficient, effective and agile organisation) of our 5-year strategy, published in March 2022, focused on strengthening our insights and analyses, to inform how we act.

Key findings

1. For the first time, data on the intersection of age, ethnicity and disability with gender is included. Breakdowns by gender and ethnicity show that White males received the largest percentage of awards as both principal investigators and co-investigators (co-investigators), receiving 57% of principal investigator awards and 49% of co-investigator awards.

2. Representation of female grant holders overall was 28% for principal investigators and 33% for co-investigators. For both role types, this is below a benchmark for the wider academic population, at 42% (for more information see the ‘benchmarks’ heading). Representation of female grant holders was above this benchmark for fellows, at 49%.

3. After the White ethnic group at 81%, the highest percentage of principal investigator awardees was from the Asian ethnic group, at 8%. The percentage of principal investigator grant holders from the Black, Mixed, and other ethnic groups was lower, at 1%, 2% and less than 0.5% respectively.

4. The award rate for principal investigators was lowest for the Black and other ethnic groups at 13% and 12% respectively. This compares with award rates of 29% for the White ethnic group, 23% for the Mixed ethnic group and 21% for the Asian ethnic group

5. The percentage of principal investigator and co-investigator grant holders reporting a known disability remains low at 2%. This is below benchmarks for wider academic staff at 4%, and the wider labour market (employed) population at 13%.

6. Disability status was not disclosed for 6% of principal investigator grant holders and 6% of co-investigator grant holders. Ethnicity was not disclosed for 8% of principal investigator awardees and 9% of co-investigator grant holders. Further investigation is needed to understand why there are higher levels of non-disclosure for disability and ethnicity than for the other legally protected characteristics, and the impact this has on our data.

7. Among UKRI’s funded postgraduate research students, representation of female students and students reporting a disability is below benchmarks for the wider postgraduate research student population. 45% of UKRI studentship starts were female compared with a benchmark of 50%, and 8% reported a disability, compared with a benchmark of 12%. A high level of ‘non-disclosure’ or ‘unknown’ for ethnicity remains for UKRI-funded studentship starts, at 32%.

Benchmarks

Benchmarks for the wider academic population are constructed from Higher Education Statistics Agency (HESA) data and include academic staff on a contract that includes ‘both teaching and research’. Both fixed-term and permanent contracts are included, as are full-time and part-time work patterns.

Contracts that are ‘research only’, ‘atypical’ or ‘dormant’ are excluded. Differences in time periods, geographic coverage and inclusion criteria mean that the HESA population is not directly comparable with the people funded by UKRI.

Figures are provided as an approximate benchmark to help with interpreting the data. Difference from the benchmark can be attributed to:

differing funding levels across the councils
the fact that not all those funded are UK based, (making the HESA benchmark less relevant)
the fact that there may be success rate differences associated with gender

Benchmarks for wider postgraduate research students are also constructed from HESA data and include doctoral students and research-based masters students. Differences in time periods, geographic coverage and inclusion criteria mean that the HESA population is not directly comparable with who UKRI funds. Figures are provided as an approximate benchmark to help with interpreting the data.

Introduction

We are committed to publishing diversity data on applications and awards against 4 diversity characteristics (age, disability, ethnicity and gender) annually, as part of our business-as-usual activities.

We are reviewing our data collection processes through development of the new funding service. Within this, we will review how data is collected on these 4 characteristics and will also widen data collection to include other protected characteristics.

The current funding services asks applicants to provide information on gender with male, female or not disclosed as response options. For this reason, we use the terminology male and female to describe gender categories throughout this document. We appreciate that gender and sex terminology are more nuanced and highly personal and will reflect this through data collection in the new funding system.

The data is broken down by role type, principal investigators, co-investigators, fellows, and studentship starts, and the data included covers 7 financial years (2014 to 2015 to 2020 to 2021). Discussion within this report is primarily focused on data for the latest financial year. Developing the capability to explore trends over time is an important next step in improving our analysis.

We publish this data to aid transparency in our funding and enable high-level monitoring of trends.

For 2020 to 2021, the data continues to reveal under-representation of females, ethnic minority groups and individuals with a declared disability when compared to the benchmark for the UK academic community and the UK workforce.

We are using this data, together with other evidence, and engagement with the research and innovation community, to support priority 6.2 of our 5-year strategy published in March 2022. This aims to strengthen our insights and analyses to inform how we act, drawing on the breadth and depth of our expertise to respond to the rapidly changing research and innovation landscape.

The 2020 to 2021 financial year was unprecedented, with the impact of the COVID-19 pandemic, and so in many ways was an atypical year for funding applications.

Our independent process review of UKRI’s response to COVID-19 revealed that we received a substantially larger volume of applications at a much faster pace than normal, with COVID-19 response applications adding around 30% to our typical annual application volume.

Success rates were also impacted, with a success rate of just over 10% for the main COVID-19 response funding opportunities, compared with an overall award rate for UKRI of 21% (although this varies substantially by opportunity and award type).

Emergency funding opportunities, set up as part of our response to COVID-19 are not included within the data in this report (COVID rapid response funding is not included. There are 3 COVID funding opportunities included in the data: MS COVID SPF 2020; SPF COVID19 Rapid Knowledge; UKRI GCRF urgency COVID funding). Separate analysis of the diversity profile for these funding opportunities is included in Annex D of the independent process review.

Early analysis of data for UKRI does not indicate substantial changes in the diversity profile of applicants and awardees because of the pandemic. However, further work including the monitoring of trends in future years, will enable us to understand the impact of the COVID-19 pandemic on funding applicants and awardees more fully.

An impact evaluation for our COVID-19 response is ongoing and will explore to what extent (and how) projects have fostered equal, diverse, and inclusive research and innovation environments.

What is included

This report is based upon competitive UKRI funding to named individuals from across the 7 research councils and pan-UKRI funds, which are centrally run schemes.

This does not reflect all of UKRI’s budget. For example, it does not include strategic funding (such as block grants to institutes), or funding from Research England and Innovate UK. The majority of Innovate UK and Research England funding goes to organisations and therefore all Innovate UK and Research England funding is excluded from this report.

Innovate UK will be publishing a report on the diversity of Innovate UK competition applicants and recipients early in 2023. In addition, a new equality, diversity and inclusion survey was introduced for applicants in May 2022 and data from this survey will be published annually.

Results are included against 4 characteristics (age, disability, ethnicity, and gender) for each of the 7 research councils, Arts and Humanities Research Council (AHRC), Biotechnology and Biological Sciences Research Council (BBSRC), Engineering and Physical Sciences Research Council (EPSRC), Economic and Social Research Council (ESRC), Medical Research Council (MRC), Natural Environment Research Council (NERC), Science and Technology Facilities Council (STFC), for pan-UKRI funds, and for UKRI as a whole.

Data is available for the 7-year period from 2014 to 2015 to 2020 to 2021. Results for UKRI’s research grants and fellowships are broken down by role type, principal investigators, co-investigators and fellows, and data is included separately for council-funded studentship starts.

For more detail on the methodology, see the methodology annex.

This year, we have made a few changes to improve the content of the report.

We have disaggregated ethnicity and included intersectionality for the first time. In the next report we will develop estimates of expected variation over time and develop methods to assess statistical significance.

External benchmark data is used to provide context and comparison to our diversity data. We use it to understand how the diversity of our applicants and awardees compares to the make-up of those eligible to apply for our funding, and the wider population.

We have used the latest data available from the Higher Education Statistics Agency (HESA) for the diversity population of the UK higher education institutes (HEIs) and the ONS Labour Force Survey, which provides the percentage of people from each ethnic group employed in the UK (see Annex A).

It should be noted that neither benchmark is a perfect comparator. Some UKRI funding opportunities allow for international applicants or researchers outside of UK HEIs, while student benchmarks for the HESA data are for UK-domiciled students only.

Discussion within this report is primarily focused upon data for the latest financial year. Changes over time are only discussed where we can provide additional evidence to support the conclusions.

Developing the capability to explore trends over time is an important next step in improving our analysis.

The following results are included:

the proportion of applicants and awardees for research grants and fellowships
award rates (number of awardees as a percentage of the number of applicants) for research grants and fellowships
mean and median application and award values for successful applicants for research grants and fellowships
award rate by value (the value of the total amount awarded as a percentage of the total amount applied for) for research grants and fellowships
the proportion of doctoral studentship starts
estimates of UK academic staff and postgraduate research student populations for each council, based on HESA data, to understand whether the applicants and awardees reflect underlying populations of staff and students in higher education

Mean and median award amounts are without indexation to enable comparison with application amounts. Results for median award amounts are discussed in this report rather than mean award amounts. This is due to the skewed nature of grant funding with some very large award amounts being awarded. These large awards can affect comparisons between groups when using the mean.

Caution must be used when comparing to HESA staff and student population. Different time periods, geographic coverage and inclusion criteria mean that the populations are not directly comparable with the people funded by UKRI. Figures are provided as an approximate benchmark to help with interpretation of the data.

Data for the cross-UKRI Future Leaders Fellowships (FLF) is included within the totals for fellows under ‘pan-UKRI funds’ in council-level breakdowns but is not separately identified. Separate data for the programme is also provided alongside this report for individual financial years. Results for the FLF programme are not discussed in this narrative, as these are already reported on and discussed separately on a round by round basis.

Intersectionality analysis is also limited to applications, awards, and award rates at this stage. We recognise that further work is needed to understand the intersection between protected characteristics more fully and will continue to develop and improve this analysis in future release.

The impact of council-level differences on UKRI-level data

Where possible, data for individual research councils is provided in accompanying data tables. Insights from analysis at council level, funding stream level or funding opportunity level are also discussed where they can inform our understanding of UKRI-level trends.

There is large variation in application numbers, award rates and award amounts across UKRI’s funding portfolio, including between research councils. These reflect differences in process and reporting methods.

The variation in award value can be due to the higher consumable and equipment costs that are necessary for some areas of research. Councils are looking into their data individually, and this substantial evidence base is directly influencing equality, diversity and inclusion workplans.

The distribution of applicants by number across councils for 2020 to 2021 is:

AHRC 3,850
BBSRC 3,585
EPSRC 7,150
ESRC 4,625
MRC 7,845
NERC 4,410
STFC 1,170
pan-UKRI 3,305

MRC has the highest number of applicants compared to the other councils and STFC the fewest.

Within this distribution of applicants there are differences in diversity characteristics and application patterns across the 7 research councils. Figure 1 shows the distribution of UKRI’s applicants across age, disability, ethnicity, and gender by the 7 research councils and pan-UKRI.

Figure 1: The number of total applicants to UKRI by diversity characteristic for the research councils (2020 t0 2021)

Figure 1 continued. The number of total applicants to UKRI by diversity characteristic for the research councils (2020 to 2021)

This highlights the fact, for example, that at UKRI level the 40 to 49 age group has the highest proportion of applicants. MRC and EPSRC have a higher number of applicants aged 40 to 49 than the other councils. EPSRC has the highest number of applicants across all age groups, and STFC and pan-UKRI funds have the lowest.

The White ethnic group has the highest number of applicants. STFC and pan-UKRI have the lowest number of applicants in this group, and Asian applicants to EPSRC are higher than both. In terms of gender, AHRC and ESRC have similar numbers of male and female applicants.

In 2020 to 2021, overall award rates for research and innovation grants ranged from 17% at MRC to 72% at STFC. Similarly, median award amounts for principal investigators ranged from approximately £135,250 at AHRC to approximately £593,536 at MRC.

At UKRI level in 2020 to 2021, the median award amount for male principal investigators was approximately £400,000, compared with approximately £300,000 for female principal investigators. However, the council-level data shows that this difference is not consistent across all the councils.

At AHRC, BBSRC, EPSRC and NERC, differences in median award amount by gender for principal investigators were smaller than at UKRI level (all the differences were less than £40,000).

Differences were larger at ESRC, MRC and STFC, where the median award amount was higher for male principal investigators than for female principal investigators.

At UKRI level, award rates were also higher for principal investigator males in 2020 to 2021 (at 29%), compared with 25% for principal investigator females.

At council level, however, the picture is mixed. Principal investigator award rates by gender were very similar (within 2 percentage points of each other) at all councils except STFC, where principal investigator females had an award rate 9 percentage points higher than males. The higher award rate for males at UKRI level is a consequence of a higher percentage of male applicants and awardees at EPSRC and STFC, which are also the 2 councils with the highest award rates.

It is therefore important to consider council-level data, alongside UKRI-level data, when comparing amounts and award rates between demographic groups.

We advise against using these findings alone to draw causal inferences regarding the relationship between the diversity characteristic and application and award rates. Further analysis is needed to control for the effects of other background factors.

Results

This section discusses key findings for the 7-year period (2014 to 2015 to 2020 to 2021). The data can be explored further using the accompanying Excel downloads and interactive dashboard.

Overall distribution in 2020 to 2021

Figure 2 shows the overall composition of UKRI applicants by age, gender, ethnicity, and disability for the three role types – principal investigators, co-investigators and fellows – in 2020 to 2021. Key findings for each characteristic are discussed in more detail below for principal investigators, co-investigators and fellows, and for studentships starts.

Figure 2: The proportion of applicants by role and characteristic, UKRI, 2020 to 2021

Principal investigator and co-investigator

Age

The age distribution of principal investigators and co-investigators has remained relatively stable over the 7-year period. For principal investigators and co-investigators in 2020 to 2021, the 40 to 49 age group continued to account for the largest share of awardees at 38% for principal investigators and 37% for co-investigators.

Differences in award rates by age group also remained stable in 2020 to 2021 if we exclude ‘unknown’. For co-investigators, there was little variance in award rate by age, ranging between 28% and 29% for all age groups. For principal investigators, award rates ranged from 20% for those aged 29 or less to 32% for those aged 50 to 59 (Figure 3).

Figure 3: Award rate for principal investigators and co-investigators by age group (UKRI, 2020 to 21)

Consistent with previous years, the median award amount for principal investigators in 2020 to 2021 increased with age, with the highest median award amount being for the 60-plus age group (£505,000), and lowest for the 29-or-less age group (£214,000), excluding the ‘unknown’ group. Award rate by value was also highest for the 60-plus age group (at 39%) and lowest for the 29-or-less age group (at 22%).

Figure 4: Median award amount by value for principal investigators by age group (UKRI, 2020 to 21)

Disaggregating award rates by age across councils is challenging due to small and suppressed values. After rounding, we have used the 5-year cumulative data set.

There is a similar range across councils and age groups. STFC has a higher award rate across all age groups for principal investigators, with the highest being at 73% for the 50 to 59 age group.

Figure 5: Award rate by value for age by councils (UKRI, 5-year cumulative)

Disability

Applications and awards

In 2020 to 2021, the percentage of principal investigator and co-investigator applicants and awardees reporting a disability was 2%. This is consistent with previous years, in which the percentage has fluctuated between 1% and 2%

It is lower than benchmarks both for HESA academic staff (at 5%) and the labour market (employed) population (at 13%). The percentage reporting a disability is also lower for all research councils than their corresponding benchmark for wider academic staff.

The percentage of applicants and awardees where disability status was not disclosed or ‘unknown’ continues to be high in 2020 to 2021 (between 6% and 8%). This is similar to that reported in relation to ethnicity (between 7% and 8%).

It is higher than the level of non-disclosure for age and gender (ranging from 0% to 2%) and for disability in the benchmark HESA data at 3%. Similar levels of non-disclosure or ‘unknown’ data occur at the council level for co-investigators, ranging from 6% at BBSRC, EPSRC and STFC to 9% at AHRC.

Further work is needed to understand both the reasons why applicants may choose not to disclose their disability status, and the nature of the barriers faced by disabled individuals.

Award rates

Differences in award rate by disability status remain consistent with previous years with a higher award rate for principal investigators and co-investigators reporting no known disability than individuals reporting a disability. In 2020 to 2021 this difference was 2 percentage points for principal investigators and 3 percentage points for co-investigators.

To compare award rates by disability status at research council level we use aggregated data over a 5-year period due to the small number of awardees.

For principal investigators and co-investigators there is a mixed picture across the research councils. For principal investigators, award rates were higher for applicants reporting a disability at AHRC, BBSRC and NERC. For co-investigators, award rates were higher for applicants reporting a disability at AHRC, ESRC and NERC, while they were on a par at MRC.

Figure 6: Award rates for principal investigators and co-investigators by disability for councils (UKRI, 5-year cumulative)

Award amounts

Figure 7 shows that in 2020 to 2021, the median award amount for principal investigators continued to be higher for applicants reporting no known disability than for applicants reporting a disability. This difference is consistent with previous years and is also reflected in the median application amounts for the 2 groups.

Award rate by value for principal investigators was also higher for applicants reporting no known disability at 30% compared with 25% for applicants reporting a disability.

Figure 7: Median award amount and award rate by value for those reporting a disability by council (UKRI, 2020 to 21)

Ethnicity

In this report, we have moved away from using binary ethnicity categories (ethnic minorities and White) and have disaggregated the data by ethnicity as much as possible. Due to the small number of awards at the level of individual ethnicities, data is primarily presented and discussed for the broad ethnic groups (Asian, Black, Mixed, other, and White).

In all years since 2014 to 2015, the highest percentage of principal investigator and co-investigator applicants and awardees were from the White ethnic group (81% of principal investigators and 75% of co-investigators in 2020 to 2021).

After the White ethnic group, the highest percentage of awardees was from the Asian ethnic group for both principal investigators and co-investigators. In 2020 to20 21, 8% of principal investigator awardees and 9% of co-investigator awardees were from the Asian ethnic group.

Representation of the Black, Mixed, and other ethnic groups remains low in 2020 to 2021 among both principal investigators and co-investigators. Representation was particularly low for principal investigators, where the percentage of awardees from the Black ethnic group was 1%, the Mixed ethnic group 2% and the other ethnic group below 0.5% (rounded down to 0%).

This low representation is also reflected in benchmark figures for the wider academic population, at 2% for each of these 3 ethnic groups.

In 2020 to 2021, ethnicity was either not disclosed or unknown for between 7% and 9% of applicants and awardees across the role types. This is higher than for the age and gender characteristics and is consistent with previous years. The HESA benchmark is 3% for not disclosed and 2% for unknown.

Figure 8 shows that the White ethnic group is the most prevalent for principal investigators (81%) and co-investigators (75%), though this is below the benchmark for the wider labour market (87%) and equivalent to the wider academic population (80%).

The representation of both principal investigators and co-investigators at 3% for the Mixed ethnic group is higher than the wider labour market (1%) and the wider academic population (2%).

The Black ethnic group is the lowest for principal investigators (1%) and co-investigators (3%), below the 3% benchmark for the wider labour market and the wider academic population (2%).

Asian co-investigators (10%) are above the benchmark for the wider academic population, and higher than the 6% of the wider labour market.

Figure 8: Percentage of awardees for the Asian, black, mixed, and other ethnic groups compared with the HESA and labour market benchmarks

Within the ethnic minority groups, trends are consistent with previous years. For principal investigators, Chinese and Indian ethnicities formed the largest share within the Asian ethnic group, while ‘Black-African’ and ‘Mixed-other’ formed the largest shares of the Black and Mixed ethnic groups respectively.

For principal investigators in 2020 to 2021, the percentage of Bangladeshi, Black and Pakistani awardees is below benchmark figures for the academic staff and labour market (employed) populations.

For co-investigators, the same is true for the Bangladeshi ethnicity. For both principal investigators and co-investigators, the percentage of awardees reporting an ‘other’ ethnicity also falls below both benchmarks.

At research council level, only 5-year cumulative figures are presented, due to the small number of awardees.

Over the 5-year period 2016 to 2017 to 2020 to 2021 for principal investigators, the percentage of awardees from the Black ethnic group ranged between less than 0.5% (EPSRC) and 2% (MRC) in every council.

There was greater variation in the percentage of awardees from the Asian ethnic group, ranging from 3% at NERC to 12% at EPSRC.

For co-investigators, the highest percentage of Black awardees were at MRC (6%). As with principal investigators, there was greater variation in the percentage of awardees from the Asian ethnic group. At AHRC 5% of awardees were from the Asian ethnic group compared with 15% at MRC.

There have been some changes in the ethnic composition of applicants and awardees over time. However, we do not currently have estimates of expected annual fluctuation and, as such, it is hard to interpret whether changes are an expected annual variation or represent true change.

Developing measures of expected variation over time, and methods to assess the statistical significance of differences between groups are an important next step for this analysis.

Some of the changes in the representation of ethnic groups over time can be explored further by looking at changes in the programmes that are funded by UKRI each year. For example, Figure 9 shows how the percentage of Asian awardees has changed since 2014 to 2015 for principal investigators and co-investigators, alongside a benchmark for the wider academic population.

Figure 9: The percentage of awardees by ethnic group, by role type and year, for co-investigators and principal investigators alongside a benchmark of the wider academic staff population (2014 to 15 to 2020 to 21)

Representation of the Asian, Black and Mixed ethnic groups is higher amongst grants classified as Official Development Assistance (ODA) for both principal investigators and co-investigators, compared with the rest of UKRI’s funding.

For example, in 2018 to 2019, the percentage of ODA co-investigator awardees from the Asian ethnic group was 19%, compared with 8% for the rest of UKRI’s funding.

For the Black and Mixed ethnic groups, representation was also higher for ODA awardees, at 7% and 8% respectively, compared with 1% and 2% for other co-investigator awards.

Some of the changes in the representation of ethnic groups over time (such as those shown in Figure 9) are due to changes in the programmes funded by UKRI. For example, the number of ODA funding opportunities fell from 62 in 2018 to 2019 to 58 in 2019 to 2020, then fell again to 29 in 2020 to 2021.

This has led to a decline in the percentage of principal investigators and co-investigators awarded grants that classify as ODA funding. In 2018 to 2019 28% of co-investigator awardees were for ODA grants: this fell to 16% in 2020 to 2021. For principal investigators, the decline was from 13% of awardees to 7% in 2020 to 2021.

If co-investigators associated with ODA funding are removed from UKRI’s data, the percentage of co-investigator awardees from the Asian ethnic group has been stable at 8% in each of the last 3 years. If principal investigators associated with ODA funding are removed from UKRI’s data, the percentage of principal investigator awardees from the Asian ethnic group was 6% in 2018 to 2019, 8% in 2019 to 2020 and 7% in 2020 to 2021.

The reduction in the number of ODA funding opportunities has not affected the ethnic diversity of applicants and awardees for all councils equally, as some councils have a higher percentage of grants classified as ODA.

AHRC has seen the largest change in the percentage of principal investigators associated with ODA funding, with a peak of 12% in 2019 to 2020, declining to 2% in 2020 to 2021.

In 2020 to 2021 the distribution by ethnic group for ODA-related grants varies by research councils. The percentage of principal investigator awardees for the Asian ethnic group was between 3% (at AHRC) and 18% (at MRC). The Mixed ethnic group ranges from 2% at pan-UKRI to 17% at NERC.

Award rates

When looking at award rates by ethnic group for principal investigators, the White ethnic group had the highest award rate in 2020 to 2021 (at 29%), followed by the Mixed ethnic group at 23% and the Asian ethnic group at 21%.

Award rates were lowest for the Black and other ethnic groups at 13% and 12% respectively. Differences in award rate by ethnic group were smaller for co-investigators, where the award rate was highest for the White and Black ethnic groups (at 29% and 28% respectively) and lowest for the other ethnic group at 21%.

Figure 10: Award rates for principal investigators and co-investigators by ethnic group (UKRI, 2020-21)

To explore award rates by ethnic group at the council level, we use aggregated figures over a 5-year period (2016 to 2017 to 2020 to 2021). This is due to the small number of awardees for some ethnic groups at this level of disaggregation.

Council-level data reveals large variation in trends. For example, MRC award rates ranged from 24% for the White ethnic group to 21% for the Asian ethnic group.

Figure 11: Award rates for principal investigators and co-investigators by ethnic group for councils (UKRI, 5-year cumulative)

For principal investigators, there are suppressed values after rounding for Black and other ethnic groups in AHRC, BBSRC, ESRC, NERC and STFC. Award rates were highest for White ethnic group. For co-investigators, award rates were higher for White ethnic group, with award rates ranging from 16% Black ethnic group at BBSRC to 61% Mixed ethnic group at STFC.

Award amounts

Due to the small number of awardees, award amounts are compared between the White ethnic group and ethnic minorities (excluding White minorities).

Consistent with previous years, principal investigators from White ethnicities continued to apply for and receive higher award values relative to ethnic minority (excluding White minority) applicants and awardees (Figure 12).

Different trends occur in award amount and award rate by value at research council level for principal investigators.

At BBSRC, ESRC and STFC in 2020 to 2021, median award amounts were higher for awardees from Asian ethnic groups than for the White ethnic group. The number of awardees that this represents ranges from 10 to 100, compared to 190 to 650 for White ethnic group across all councils.

Disaggregating award amounts at the council level by principal investigator is challenging for Black, Mixed, and other ethnic groups as the values are small, with suppressed values after rounding.

The award amounts vary across the councils and by ethnic groups. The median value of awards for principal investigators ranges from £120,000 for the mixed ethnic group at AHRC to £1,735,000 at MRC.

Figure 12: Median award amount and award rate by value for principal investigator ethnicity (UKRI, 2020 to 21)

Gender

The percentage of both applicants and awardees reporting their gender as female has steadily increased for the co-investigator role type by 8 percentage points since 2014 to 2015.

For principal investigators, the trend is less clear and, while the percentage of female principal investigator applications has increased steadily (plus 5 percentage points overall), the percentage of female awards has fluctuated.

Despite some improvements in female representation for each role type, representation was below a benchmark for the wider academic population in all the last 4 years.

In 2020 to 2021, the benchmark for female representation was 42% compared with 33% for co-investigator awards and 28% for principal investigator awards.

Figure 13: The percentage of female applicants and awardees by role type and year for principal investigators and co-investigators alongside a benchmark for the wider academic population (2014 to 15 to 2020 to 21)

Trends in applications and awards by gender continue to vary considerably at research council level.

For principal investigators, AHRC and ESRC had the highest percentage of female applicants (48% and 49%) and awardees (49% and 51%) in 2020 to 2021, their respective HESA benchmark being 48% for AHRC and 51% for ESRC.

For all other councils, the percentage of principal investigator female applicants was below their HESA benchmark, ranging from BBSRC (with 26% of female awardees and applicants, compared to the benchmark of 40%) to MRC (with 36% and 41%).

Figure 14: The percentage of female applicants and awardees for principal investigators by research council, alongside a benchmark for the wider academic population (2020 to 21)

Award rates

For UKRI overall, the proportion of grants awarded to principal investigators and co-investigators was higher for males in 2020 to 2021, with 70% for principal investigator and 64% for co-investigator, while the proportion of grants awarded to females overall was 28% for principal investigators and 33% for co-investigators.

However, disaggregating data for principal investigators reveals that for most councils, award rates for female principal investigators are greater than or equal to those of male principal investigators. The only exceptions are MRC (minus 1 percentage point) and NERC (minus 2 percentage points), which account for about 33% of all awardees.

Looking at award rates of co-investigators by gender, we find that the award rates are similar for all councils. The difference in favour of male co-investigators is in the range of 0.5 percentage points to 1 percentage point when disaggregated by council.

These trends vary at research council level where for principal investigators, higher award rates occurred for females at AHRC, ESRC and STFC and for female co-investigators at AHRC, EPSRC and STFC.

Figure 15: Award rate by gender for each council (UKRI, 2020 to 21)

Award amounts

In 2020 to 2021 for principal investigators, median application and award amounts for females continued to be lower than for males.

The median application amount for male principal investigators in 2020 to 2021 was 20% higher than for females and the median award amount was 28% higher.

This difference partly reflects differences in the average award amounts of councils and the percentage of male and female principal investigators who are funded by each council.

For example, there is a higher percentage of female principal investigators funded by ESRC and AHRC, where average award amounts are lower than in some other councils.

Award rate by value for male principal investigators was also higher (at 30% compared with 25% for females Figure 16).

These trends vary at research council level, with higher median application and award amounts for female principal investigators funded by BBSRC, EPSRC and NERC.

The difference in the median application and award amounts is small: £1,000 for BBSRC and £3,000 for EPSRC. AHRC was the only research council in 2020 to 2021 with a higher award rate by value for female principal investigators, while STFC had gender parity for principal investigators in award rates by value.

Figure 16: Median award amount and award rate by value for gender, by councils (UKRI, 2020-21)

Intersectionality with gender for principal and co-investigators

For the first time, this report includes data on the intersection of age, disability, and ethnic group with gender.

Data is currently limited to 1 financial year (2020 to 2021), however, and so developing our intersectional analysis to include trends over time is an important next step.

We have initially focused on the intersection of the characteristics with gender, due to the small number of applicants and awardees in some of these other groups.

Figure 17 shows the composition of awardees for principal investigators and co-investigators in 2020 to 2021. For principal investigators, the largest group of awardees were white males, receiving 57% of awards, followed by white females, receiving 24% of awards.

The next largest group was Asian males with 6% of awards. The remaining groups each received between less than 0.5% (rounded down to 0%) and 1% of awards. Gender or ethnic group was either unknown or not disclosed for the remaining 7% of awardees.

The composition of co-investigator awardees was similar, with the highest percentage of awards received by white males (49%), white females (25%) and Asian males (6%).

Representation of Asian females amongst co-investigator awards was higher (at 3%) than for principal investigator awards. Similarly, although still low overall, representation of Black males (2%), Black females (1%) and males reporting a Mixed ethnicity (2%) were higher for the co-investigator than for the principal investigator role.

Figure 17: The composition of principal investigator (PIs) and co-investigator (CIs) awardees by ethnic group and gender (2020-21)

Intersection of age and gender

In 2020 to 2021, the highest proportion of principal investigator and co-investigator applications and awards was from males aged 40 to 49. For principal investigators, this was 24% of all awardees and for co-investigators 23% of all awardees.

For all age groups between age 30 and 60, the percentage of principal investigator and co-investigator awardees was higher for males. For both males and females there was little difference in the small percentage of awardees aged 29 or less.

For principal investigators and co-investigators, the award rate for males is higher than for females for all age groups except co-investigators aged 60-plus where award rate for females was 29% compared to 28% for male.

The largest difference in award rate by gender and age was for the co-investigator age group 29 or less, where the award rate for males was 5 percentage points higher than for females (at 30% and 25% respectively; the overall number of applicants in this group is low).

Figure 18: Award rate by age group and gender for principal investigators and co-investigators (UKRI, 2020 to 21)

Intersection of gender and disability

For principal investigators and co-investigators, 1% of awardees in 2020 to 2021 were females reporting a disability and 1% were males reporting a disability. As males receive a higher proportion of principal investigator and co-investigator awards overall, this means that a higher proportion of female awardees reported a disability than male awardees. Only 26% of principal investigator awardees were females reporting no known disability compared with 65% of males.

For principal investigators in 2020 to 2021, male applicants reporting no known disability had the same award rates as males reporting a disability (30%). Female applicants reporting no known disability had a higher award rate than female applicants reporting a disability. Males also had a higher award rate than females regardless of disability status. For co-investigators, the award rate for females with no known disability was 27% compared with 23% for females reporting a disability.

Figure 19: Award rate by gender and disability status for principal investigators and co-investigators (UKRI, 2020 to 21)

Intersection of ethnicity and gender

For application and award proportions, differences by gender and ethnicity were largest for the principal investigator role.

In 2020 to 2021, White males received 57% of principal investigator awards, compared with 24% that were awarded to White females.

The next largest difference by gender was for the Asian ethnic group with Asian males receiving 6% of principal investigator awards, while only 1% were awarded to Asian females. Proportions from the Black, Mixed and other ethnic groups were similar in terms of the gender of principal investigators.

For co-investigators, the highest proportion of awardees in 2020 to 2021 was also among White males, at 49% compared with 25% that were awarded to White females. The next largest proportion of awards went to Asian males at 6%. Females from the Black and Mixed ethnic groups received 1% of co-investigator awards each.

As mentioned above, it is preferable to use 5-year cumulative figures when comparing award rates between ethnic groups due to the small number of awardees within some groups.

However, 5-year figures are not yet available for the intersection of ethnicity and gender so annual results for 2020 to 2021 are presented. These results should therefore be interpreted with caution. Additional data will be needed in future years to assess the robustness of these findings.

In 2020 to 2021, female co-investigators from the Black ethnic group had the highest award rate of all ethnic groups and genders, at 33%.

This trend was not observed for principal investigators, where Black females had one of the lowest award rates (at 16%). This was higher only than those for Black males and Asian females (at 12% each).

For principal investigators, White males and White females had the highest award rates at 30% and 28% respectively. The Asian ethnic group had the largest difference in award rate by gender for principal investigators, at 12% for Asian females and 25% for Asian males.

Figure 20: Award rate by ethnic group and gender for principal investigators and co-investigators (UKRI, 2020 to 21)

Fellows

The composition of UKRI’s fellows differs from the composition of other role types, with greater representation of females and individuals reporting a disability. For example, in 2020 to 2021, 4% of fellow awardees reported a disability, compared with 2% of principal investigator and co-investigator awardees. Additionally, 39% of fellow awardees were female, compared with 31% of principal investigators and 35% of co-investigators.

The age distribution of fellow awardees also differs substantially from principal investigators and co-investigators, with a much higher percentage of awardees being aged below 40. In 2020 to 2021, 73% of fellow awardees were aged below 40 compared with 24% of principal investigators and 23% of co-investigators.

Fellowships are typically targeted at researchers earlier in their careers, as they seek to transition to an academic role, and this is a factor in the age demographic that we observed.

For fellows in 2020 to 2021, there were some large differences in award rates and award amounts for different groups. However, the number of UKRI fellows is much lower than with other role types (approximately 525 fellowships were awarded in 2020 to 2021 compared with thousands of principal investigator and co-investigator awards).

Therefore, award amounts and award rates for fellows are more susceptible to year-on-year changes in individual funding opportunities.

One funding opportunity that has a large impact on UKRI’s fellowship data is ESRC’s postdoctoral fellowship funding opportunity, which accounted for 24% of all UKRI fellowships awarded in 2020 to 2021.

ESRC’s postdoctoral fellowships differ from other fellowship opportunities provided by UKRI. The fellowships run for only 1 year, and they have a capped award amount of approximately £100,000.

We also do not hold data for unsuccessful applicants for these fellowships and, as such, the award rate artificially appears as 100% in our data. The characteristics of this opportunity can, therefore, affect UKRI-level data in 3 ways by:

lowering average award amounts
increasing award rates
increasing award rates by value

Data for the cross-UKRI Future Leaders Fellowships (FLF) is included in the totals for fellows under ‘pan-UKRI funds’ in council-level breakdowns, but is not separately identified.

Separate data for the UKRI FLF is also provided alongside this report for individual financial years. Results for the Future Leaders Fellowships are not discussed in this narrative as these are already reported on and discussed separately on a round-by-round basis.

Age

In 2020 to 2021, variation in award rates by age for fellows ranged from 22% for 30 to 39 year-olds to 29% for fellows aged 60 plus.

Median award amounts for fellows were highest for the 40 to 49 and 30 to 39 age groups and lowest for the 29-or-less age group.

The median award amount for 40 to 49 year-olds was £942,000 compared with £102,000 for the 29-or-less age group.

This difference is affected by the types of fellowship that different age groups were awarded. For example, 64% of fellows aged 29 –or less were ESRC postdoctoral fellows, for which award amount was capped.

Figure 21: Award rate and median award amount for fellows, by age group (UKRI, 2020 to 21)