By Vikki Boliver

The Runnymede Trust has just launched its publication Aiming Higher: Race, Inequality and Diversity in the Academy which shines a spotlight on ethnic inequalities in UK universities. The report brings together 15 short essays written by academics and policy makers which make clear that radical change is needed to address ethnic inequalities in university admissions, student experiences, degree attainments, graduate labour market outcomes, and access to academic positions especially at senior levels.

In my contribution to the Runnymede publication (see chapter 5) I focus on the issue of ethnic inequalities in university admissions chances. Although British ethnic minorities are more likely to go to university than their White British peers, some ethnic minority groups – notably the Black Caribbean, Black African, Pakistani and Bangladeshi groups – remain strikingly underrepresented in the UK’s most academically selective institutions including Russell Group universities. Of course this is partly due to ethnic inequalities in secondary school attainment which means that members of these groups are less likely to have the high grades required for entry to highly selective universities. But we also know, from analysing university admissions data that British ethnic minority applicants are less likely to be offered places at highly selective universities even when they have the same grades and ‘facilitating subjects’ at A-level as White British applicants. Offer rates from Russell Group universities are 3 to 16 percentage points lower for British ethnic minority applicants than for white British applicants after differences in A-level attainment have been taken into account, as the table below shows. This applies to applicants of Chinese, Indian and Mixed ethnicity too – these typically high achieving ethnic minority groups are not under-represented in Russell Group universities, but this is despite the fact that they are less likely to get in than comparably qualified white applicants.

Percentage of applications met with an offer of admission from Russell Group, other Old, and New universities after taking attainment at A-level into account
	Russell Group universities	other Olduniversities	Newuniversities
White British	52.0~	62.1~	60.8~
Black Caribbean	44.7*	57.8*	56.7*
Black African	35.7*	50.6*	50.7*
Pakistani	39.6*	58.3*	58.5*
Bangladeshi	42.6*	57.9*	62.3~
Indian	47.6*	61.9~	60.9~
Chinese	48.4*	62.0~	59.7~
Mixed	48.8*	59.7*	59.0*
Other	44.2*	59.7*	58.4*
Note: An asterisk (*) indicates that the figure is statistically significantly different at the p. < 0.05 level from the corresponding figure for the white British group.Source: Aiming Higher p.16

It is also clear from analysing admissions data that this pattern of lower offer rates for ethnic  minority applicants is not confined to the Russell Group but is in fact evident for other Old and New universities as well – although the ethnic gap in offer rates is rather smaller at 3 to 4 percentage points. What this indicates is that ethnic inequalities in university admissions are not peculiar to elite institutions – this is in fact a sector wide problem.

So what can be done about this? I want to suggest two things that could be done, or at least started, today.

Firstly, universities clearly need to be encouraged, and if necessary required, to undertake searching reviews of their admissions policies and practices. Universities must seriously consider whether some of the criteria and processes being used to select students systematically and unwarrantedly disadvantage ethnic minority applicants. For example, universities often use information about applicants’ predicted grades at A-level (and its equivalents) when making admissions decisions. This practice may unfairly disadvantage ethnic minority applicants if their grades are more likely to be under-predicted than white applicants’ grades. (We don’t know whether this is the case, but UCAS holds the data needed to find out.) Universities must also take seriously the possibility that admissions selectors are influenced by unconscious bias – stereotypes about different ethnic groups, social classes, genders, and so on, that are prevalent in our society and which we all pick up on whether we wish to or not. We risk falling back on these unconscious biases when we make decisions without clear criteria and procedures especially when under time pressures. The Equality Challenge Unit has developed training packs designed to help university staff recognise and resist unconscious bias. Universities would do well to extend this training to all staff involved in the recruitment and selection, and indeed teaching, of students.

Secondly, universities need to commit collectively to making their applications and admissions data open to analysis and reanalysis by external researchers and to having the findings of this research published and discussed openly. The ethnic disparities in offer rates that I have reported here are based on my analysis of anonymised individual-level data supplied by the Universities and Colleges Admissions Service (UCAS). But of late UCAS has decided, presumably with the agreement of the universities that subscribe to its services, that individual-level applications and admissions data will no longer be shared with external researchers. This is surely unacceptable. Universities are public institutions after all, they receive large sums of public money and they have a legal duty to comply with and promote the aims of the 2010 Equalities Act. Health service institutions routinely supply anonymised individual-level patient data to researchers and regulators, why should universities be any different? Individual-level university admissions data needs to be made open access not only to ensure transparency and public accountability, but also to enable multiple independent data analysts to test various hypotheses about the reason for the ethnic gap in offer rates. I mentioned earlier that UCAS holds the data needed to find out whether ethnic minority universities applicants are being disadvantaged by overly conservative A-level grade predictions. UCAS also holds the data needed to test the Russell Group’s often repeated but as yet unevidenced assertion that ethnic and school type differences in offer rates for applicants occur because “many good students haven’t taken the subjects needed for entry and universities need students not only to have good grades but grades in the right subjects for the course they want to apply for”. This may be wholly true, partly true, or not true at all. But without rigorous independent analysis of the data and detailed publication of the findings it cannot and must not be taken as fact.

The time has come for universities to rigorously review their admissions practices and to make their admissions data open data. But it would be unrealistic to expect any one university to volunteer to be the first to put their head above the parapet. Clearly universities need to come together as a sector to acknowledge and tackle what is, after all, a sector wide problem. There is considerable scope for the Office for Fair Access (OFFA), the Supporting Professionalism in Admissions programme (SPA), the higher education funding councils, Universities UK, and specific university mission groups, to bring universities together around these issues. Universities must take collective action on ethnic inequalities not just in relation to offer rates but throughout the academy.

 Vikki Boliver is Lecturer in Sociology at the School of Applied Social Sciences, Durham University