Evaluating the impact of the race equality charter on diversity staff outcomes

ABSTRACT

The Race Equality Charter (REC) was established with the main objective of tackling persistent racial inequalities in the UK Higher Education sector. There are currently over 100 REC member institutions holding close to 50 awards between them. While the sector uptake of the REC is notable, there are currently no empirical studies evaluating its impact. Using data from the Higher Education Statistics Agency (HESA), this study uses a matched difference-in-difference approach to investigate the effect of the REC on the proportion of staff from Black, Asian, and minority ethnic groups, their corresponding share of total staff in professorial or senior academic posts, as well as their corresponding share of total staff holding permanent contracts. The results reveal non-significant treatment effects which indicates that there is no difference in diversity staff outcomes between REC and non-REC institutions.

KEYWORDS:

• Racial inequalities
• Higher Education
• race equality charter
• diversity and inclusion

1. Introduction

The lack of diversity in the UK Higher Education (HE) sector has been an important policy concern for decades. Although students from Black, Asian, and minority backgrounds are now well represented relative to their share of the population, staff diversity is still lagging behind. Moreover, staff from Black, Asian, and minority backgrounds are also much less likely to be in senior positions or holding permanent posts (Bhopal 2016).

Racial inequalities in HE are evident in the low proportions of students from minority ethnic groups in the most prestigious institutions (the most selective universities), despite similar rates of application by White applicants and those from minority ethnic groups (Boliver 2013). They are also much less likely to be awarded a first class or upper second-class degree, relative to their White peers (Office for Students, office 2020). In addition, students from minority ethnic groups often report a feeling of unease, or not being able to enjoy the same levels of entitlement as their peers from White backgrounds which has consequences for their academic attainment and wellbeing (Stuart et al. 2011). Black academic staff are also likely to experience some form of microagression or passive bullying throughout their academic career (Rollock 2021).

In response to these racial inequalities, the Race Equality Charter (REC) was launched in 2015 by the Equality Challenge Unit (ECU) as a pilot with 21 institutions. The REC is guided by five main principles which emphasise the prevalence of racial inequalities in higher education institutions, the need to recognise that individuals from Black, Asian, and minority ethnic groups are a diverse group with varying experiences and outcomes, as well as the importance of working toward long-term institutional and culture change.

To become a member of the charter, the vice Chancellor/Principal of an institution must submit a letter of commitment to upholding the guiding principles of the charter. Institutions which join the charter are committed to prioritising race equality and adhering to the charter’s guiding principles. The REC also encourages members to submit an application for a bronze award within 3 years of joining the charter. A bronze award signifies that an institution is committed to tackling racial inequalities and have developed a plan of action to be implemented. To be considered eligible for a bronze award, an institution must form a self-assessment team (SAT) which must be chaired by a senior official at Vice-Chancellor level or similar. Institutions must also administer a survey to staff and students from minority ethnic groups to gain better understanding of the institutional culture, analyse the results, and engage in consultations with the target group on challenges relating to race equality. Information gathered from the surveys and consultations must be prepared in an application form to be submitted along with an evidence-based plan to address these issues (REC 2021).

Institutions who have received a bronze award in the past may also apply for a bronze renewal or a silver award. A bronze renewal is awarded to institutions which can evidence their commitment to tackling racial inequalities but still have to make some progress before being eligible for a silver award. To be eligible for a bronze renewal, an institution must show evidence of successful steps towards their action plan. For instance, demonstrating accountability through greater faculty involvement and visible public commitment from faculty since the last application (REC 2021).

To receive a silver award, an institution must be able to demonstrate significant achievements made towards eliminating racial inequalities. Evidence should be institution-wide and may include progress shown by individual faculties in implementing activities to address race equality. Institutions should also be able to evidence reflective learning which is useful for further improvement. Any institutions that fail to apply for an award within 5 years of joining the charter will have their membership withdrawn (REC 2021).

The 3-year pilot of the REC which ran between 2015 and 2018 involved 21 participating institutions, 8 of which were awarded bronze awards. At the time of writing this paper, there are now 98 REC members holding 38 bronze awards between them. There are currently no silver award holders.

A survey conducted by Oloyede, Christoffersen, and Cornish (2021) found that some of the main drivers for institutions joining the charter were the ethnicity awarding gap, the lack of progression of staff from Black, Asian, and minority ethnic groups, and the lack of representation of staff from Black, Asian, and minority ethnic groups at senior levels. Further to REC membership, Rollock (2018), notes that institutions who were unsuccessful in obtaining REC awards were unable to detail the racial inequalities faced by staff and students from Black, Asian, and minority backgrounds at their institutions, and institutions which had difficulty showing a concrete plan of action towards tackling racial inequalities also tended to be unsuccessful in obtaining REC awards.

Membership of the charter has been idealised as one of the most effective ways of reducing racial bias in higher education (Arday and Safia Mirza 2018). However, there are very few empirical studies evaluating the impact of the charter. Thus, this study contributes to the limited literature on this topic by employing a quasi-experimental approach to evaluating how effective the REC is in improving the representation of staff from Black, Asian, and minority ethnic groups overall, and at the most senior levels within institutions (i.e. professors or other senior academic positions). Therefore, this study investigates the following research questions:

• H1: Does the REC lead to an increase in the share of staff from Black, Asian, and minority ethnic groups?

• H2: Does the REC lead to an increase in the share of staff from Black, Asian, and minority ethnic groups in professorial or other senior academic positions?

• H3: Does the REC lead to an increase in the share of staff from Black, Asian, and minority ethnic groups in permanent positions?

2. Data

The data used for this study is obtained from the Higher Education Statistics Agency (HESA) and covers the academic years 2014/15 to 2015/16 as the pre-intervention years and the 2021/22 academic year which marks the post-intervention period.

The population of interest are staff who report their ethnicity as Black, Asian, mixed or other. These numbers are summed up per institution each year and then an average is calculated separately for REC institutions which make up the intervention (I) institutions and non-REC members which form the comparator (C) group. The outcomes of interest are the share of all staff from Black, Asian, and minority backgrounds in REC institutions, the proportion from minority groups who are professors, or in other senior academic positions, as well as the proportion of permanent staff from Black, Asian, and minority backgrounds.

Table 1 below displays the share of individuals from Black, Asian, and minority backgrounds who count towards total staff, total professors and senior academic staff, and total permanent staff. In all three cases, the shares of staff from Black, Asian, and minority backgrounds is higher for REC members relative to non-REC members with a slight increase in the two groups between 2014/15 and 2015/16, followed by a more modest increase in 2021/22.

Table 1. Share of individuals from black, Asian, and minority backgrounds in intervention and comparator institutions.

	2014/15		2015/16		2021/22
	I	C	I	C	I	C
% total staff	11.80	8.42	12.15	8.48	16.56	14.88
% professor or senior academic positions	3.34	1.80	3.47	1.79	4.17	1.83
% permanent staff	8.14	6.24	8.57	6.44	12.03	10.28
N	3,379	1,101	3,342	1,141	3,763	1,449

*I: Intervention institutions.

*C: Comparator institutions.

3. Method

In order to determine the causal impact of REC membership on staff diversity outcomes, a true experiment which randomises institutions into the intervention group (REC membership) and a comparator group of non-REC members are required. However, given that randomisation is not feasible in a case like this, and institutions are instead able to self-select into REC membership, this study instead applies a difference-in-difference (DiD) approach, which is a quasi-experimental research design that observes the outcomes of the intervention and comparator group over time. For the DiD to be possible, the outcomes for the two groups need to be observed prior to the intervention (in this case, the launch of the REC), and these outcomes should follow similar trends in the pre-intervention period in order for any changes in these outcomes in the post-intervention period to be attributed to the intervention.

Another important consideration for the DiD is that the comparator group should be as similar as possible to the intervention group. Given that the intervention group consists of institutions with REC membership, using all non-REC members as the comparator group may lead to biased estimates given the possible levels of variation between institutions. Therefore, we instead use a comparator group obtained via a matching process based on the following institutional characteristics; number of students (as a measure of size), Russell group membership,¹ and type of institution² (as measures of institution type), and share of female staff (as a measure of diversity) which resulted in 30 comparator institutions matched with 98 intervention institutions. To match the intervention group of REC members with the most suitable comparator group of non-REC members, a one-to-many nearest neighbor matching process with mahalanobis distance (the most conventional method for matching without propensity scores) which is based on correlations between variables was used. The aim was to match each REC institution with the closest non-REC institution based on the characteristics identified above, allowing each suitable comparator institution to serve as a match for multiple REC institutions. Therefore, the match that led to the best balance (observed by lower mean differences between intervention and comparator institutions based on the institutional characteristics identified above) was chosen.

A key consideration is that the intervention group includes 21 institutions which participated in the pilot phase of the REC and as such may attenuate the effect of the intervention when included in the sample. Thus, to account for this, the analysis is carried out in two stages; the first stage includes the pilot institutions while the second stage excludes them to observe whether there are any differences in the treatment effect. The 30 identified comparator institutions were maintained in both stages of the analysis.

Figures 1-3 below depict the parallel trends for the three outcomes of interest in the pre-intervention period. The intervention institutions represent all REC members while the comparator institutions are the matched non-REC members.

Figure 1. Share of staff from Black, Asian, and minority backgrounds.

Figure 2. Share of staff from Black, Asian, and minority backgrounds in professorial/senior academic roles.

Figure 3. Share of staff from Black, Asian, and minority backgrounds in permanent positions.

Figures 1-3 in above show that prior to the intervention, both the comparator and intervention institutions followed similar trends (slight increase over time), this means that we can safely assume that in the absence of the intervention (the REC), the two groups would continue to follow the same trends over time which allows us to attribute any changes in the post-intervention period to the REC, using the DiD approach.

To assess the quality of the matching process, the standardised mean differences between intervention and comparator institutions are calculated before and after matching on key characteristics (shown in Table 2). The results show that the matching exercise was successful in minimising the variation between the two groups based on observable characteristics such as number of students, size, type, and diversity.

Table 2. Means and standardised mean differences between intervention and comparator institutions.

	All Data			Matched Data
	I	C	(I-C)	(I-C)
	Means	Means	Std. mean diff	Std. mean diff
Number of students	18,750.97	6,906.08	0.85	0.01
Share of female staff	56.13	59.58	−0.50	−0.25
Russell group	0.21	0.05	0.41	0.32
Large universities	0.28	0.03	0.55	0.00
Mid-size universities	0.09	0.11	−0.07	0.00
Smaller universities	0.15	0.03	0.34	0.00
Very large, very high research-intensive universities	0.07	0.16	−0.34	0.00
Large high research-intensive universities	0.18	0.03	0.39	0.00
STEM-focused universities	0.02	0.09	−0.48	0.00
Arts-focused universities	0.01	0.18	−1.65	0.00
N	3,379	1,101	3,342	1,141

*I: Intervention institutions.

*C: Comparator institutions.

While existing qualitative reviews of the REC (Oloyede, Christoffersen, and Cornish 2021; Rollock 2018) have significantly contributed towards understanding the barriers to implementation of the REC in the early stages of its adoption by the HE sector, applying a quantitative approach to the evaluation goes a step further by making it possible to estimate the causal impact of the REC on the outcomes for staff from Black, Asian, and minority ethnic backgrounds. This approach is relevant for answering the identified research questions in this study as the focus of this evaluation is to determine whether REC membership has had an impact on the numbers of staff from minority ethnic groups overall, in permanent positions, and in professorial or senior academic positions. Additionally, by using a quasi-experimental design that incorporates matching on institutional characteristics, this limits bias by ensuring that we are only comparing institutions with similar characteristics to each other.

However, using a quantitative approach also has its drawbacks. One major limitation is the small sample sizes for academic staff from smaller ethnic subgroups, which means that it is not possible to investigate the impact of the REC on individual ethnic subgroups e.g. Asian -Bangladeshi or Asian – Pakistani. Given that the focus of the REC is on addressing racial inequalities that affect all staff and students from Black, Asian, and minority ethnic groups overall, using broad ethnic groups works in the context of this evaluation. However, for studies with a focus on a particular ethnic subgroup, this could distort the results. For instance, the Black ethnic group includes Black – British, Black – the Caribbean, and Black – African subgroups which are completely different cohorts with potentially diverse experiences. As such, reporting results for the broad ethnic rather than the subgroup of interest may confound the impact analysis.

4. Results and discussion

The results from the DiD regression models are presented in Table 3. The first model (1) includes the institutions which participated in the REC pilot program while the second model (2) excludes these. Including the pilot institutions within the model may attenuate the effect of the treatment while excluding them is likely to mark up the treatment effect. The relationships between REC membership and the outcomes of interest are estimated while controlling for institution type and Russell group membership.

Table 3. Difference-in-difference estimates of the impact of REC membership on outcomes for staff from black, Asian, and minority ethnic groups.

	(1)			(2)
	With pilot institutions			Without pilot institutions
Outcomes	ß	SE	R^2	ß	SE	R^2
% total staff	1.86	2.52	0.22	1.54	2.62	0.24
% professor or senior academic positions	1.12	0.93	0.25	1.24	0.97	0.27
% permanent staff	1.73	2.07	0.23	1.49	2.15	0.24

Abbreviation: SE, standard error.

4.1. Overall staff diversity

As shown in Table 3, there were no significant differences in the share of staff from Black, Asian, and minority backgrounds between REC and non-REC institutions in the 2021/22 academic year. More specifically, there is a difference of approximately 2% points (B = 1.86, p > 0.05) (H1 rejected) in the proportion of total staff from Black, Asian, and minority ethnic groups between REC member institutions and non-REC members. When pilot institutions were excluded as in model 2, the treatment effect reduces slightly although the coefficient remains nonsignificant.

The coefficient on the time variable (not shown) is positive (B = 3.27, p > 0.05) and represents the change over time, independent of the intervention. This shows that overall, there has been a positive change in staff diversity, in line with data from the Equality Challenge Unit (2022) Statistical Report which showed a gradual increase in the proportion of all staff from Black, Asian, and minority ethnic group, although this growth has been more evident among professional services staff, rather than academic staff (Equality Challenge Unit 2022). Therefore, while there has been a small improvement in staff diversity in the UK Higher Educations sector, there is no evidence that the REC itself has independently led to an improvement in staff diversity among member institutions.

4.2. Occupational group

The most recent Equality Challenge Unit (2022) report reveals relatively small difference between the share of White professors and the share of professors from Black, Asian, and minority ethnic groups. However, there were major differences within the composition of the latter where for example, there were reportedly almost 5 times more Chinese professors than Black professors. Moreover, the proportion of senior academic positions held by staff members from Black, Asian, and minority ethnic groups was roughly 50% less when compared to White senior academics.

Staff from Black, Asian, and minority ethnic groups tend to be over represented at the lowest contract levels e.g. as support staff and research assistants while they remain underrepresented at the highest levels, e.g. heads of institutions (Bhopal and Henderson 2021). The results shown in Table 3 confirm that there were no significant differences in the share of staff from Black, Asian, and minority backgrounds in professorial or other senior academic positions between REC and non-REC institutions. More specifically, there is a non-significant difference of approximately 1% point (B = 1.12, p > 0.05) (H2 rejected) in the proportion of total staff from Black, Asian, and minority ethnic groups in professorial or senior academic positions between REC member institutions and non-REC members. When pilot institutions were excluded as in model 2, the treatment effect actually increases slightly although the coefficient remains nonsignificant.

Arday and Safia Mirza (2018) find that staff from Black, Asian, and minority backgrounds are much less likely to be professorial appointments, in senior academic positions or other senior decision-making roles due to various forms of racism that reduces their chances of being selected for promotions. These results are even more pronounced for women from minority backgrounds (Singh and Kwhali 2015), particularly for those in more elite institutions (Jones 2006).

4.3. Contract type

Temporary, fixed- term academic contracts have traditionally been used for staff members whose primary role is research. However, its use has increasingly expanded towards teaching roles as well (Bryson 2004). By definition, fixed-term contracts are less secure, and given that staff members from Black, Asian, and minority ethnic groups are more likely to hold fixed-term contracts, this puts them at a disadvantage when it comes to career progression, which can exacerbate the racial inequalities they experience (Pilkington 2020). Therefore, while overall staff diversity may be improving slightly, the proportion of staff from from Black, Asian, and minority ethnic groups holding fixed-term contracts is also increasing relative to White academic staff Equality Challenge Unit (2022).

One of the main objectives of the REC is progression of staff from Black, Asian, and minority ethnic groups and this can be measured by an increase in the number of diversity staff holding permanent contracts. Furthermore, the results in Table 3 show a non-significant difference of approximately 2% points (B = 1.73, p > 0.05) (H1 rejected) in the proportion of staff from Black, Asian, and minority ethnic groups holding permanent contracts between REC member institutions and non-REC members. However, the coefficient on the time variable (not shown) is positive but not significant (B = 2.57, p > 0.05) which shows that there has been no significant change in the share of staff from Black, Asian, and minority holding permanent academic contracts over time, independent of the intervention.

5. Conclusion

To our knowledge, this study is the first to apply a quasi-experimental approach to evaluate the impact of the REC on diversity staff outcomes by comparing the outcomes of non-REC institutions. In recent years, there has been a steady expansion in the use of fixed-term contracts in the UK Higher Education sector, for both teaching and research purposes (Fontinha, Van Laar, and Easton 2018). There has also been a shift towards fixed term positions, even for positions which were originally designed to be permanent posts (Barnes and O’Hara 1999). However, staff from minority ethnic backgrounds tend to be disproportionately represented in these fixed – term positions. The findings from this study indicate that although there have been minor improvements in the outcomes of staff from Black, Asian, and minority ethnic groups since the REC was established, there is no evidence to claim that these changes have been due to the charter, given that similar trends have also occurred in the comparator group.

There were no significant differences in the share of diversity staff members overall, the proportion of diversity staff members in professorial or senior academic positions, or holding permanent contracts between REC members and non-members. Of the 99 REC members, almost 40% have achieved a bronze award which is an indication that real efforts are being made to reduce identified racial gaps. A study by Bhopal (2020) highlights that in order for the charter to make significant impacts in improving outcomes of diversity staff, sustainable, cultural changes must be made to prevent it from simply being another item on a check-list. Additionally, Singh and Kwhali (2015) note that many institutions still perceive racial inequality as an inconvenient necessity rather than a priority. Bhopal and Pitkin (2018) also note that, for real institutional change to take place, there needs to be institution-wide adoption of REC principles which requires continued efforts with impacts observed on a long-term basis. Therefore, a longer-term study is required in order to determine substantial impacts of the REC on diversity staff outcomes.

One of the guiding principles of the REC is the consideration that staff from Black, Asian, and minority ethnic staff and students are not a homogeneous group, and tend to face varying backgrounds and outcomes in higher education. For instance, the small increase in diversity staff numbers has been uneven, with staff from the Caribbean, Pakistani, and Bangladeshi origins being underrepresented in higher education, and other ethnic groups reporting high levels of racism and discrimination (Jones 2006). However, the small numbers of staff from individual minority ethnic groups, especially Black staff members, makes it very difficult to conduct a quantitative analysis of outcomes for particular ethnic groups. Therefore, much more work needs to be done in the recruitment of diversity staff members overall, and more importantly, from the individual ethnic groups that are most underrepresented in order to carry out more robust empirical analysis on the impact of the REC in reducing racial inequality.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. The Russell group consists of 24 leading UK universities, producing almost 70% of the UK’s world-leading research. See more here: https://russellgroup.ac.uk/about.

2. This is based on a cluster analysis by UK Research and Innovation (UKRI) which categorises HE institutions by their level of research and areas of focus, as broad discipline – based HE institutions or specialist HE institutions. See more here: https://kef.ac.uk/about.