The evidence behind name-blind university applications

A few weeks ago - in an article for the Sunday Times – the Prime Minister lamented that a young black man was more likely to be in prison than to be studying at a top British university. The Government intends to tackle what it deems “ingrained, institutional and insidious” discrimination at elite universities.

So, the PM has announced that UCAS will introduce ‘name-blind applications’ – whereby university applications are anonymised – from 2017. The theory is that name-blind applications will reduce “unconscious bias”, in which universities unknowingly discriminate against applicants with ‘foreign sounding’ names. Ultimately, this will lead to an increase the number of individuals from black, minority and ethnic backgrounds (BME) attending university.

But what is the evidence on the effectiveness of name-blind applications?

In universities

Very few studies have examined the effect of name-blind applications specifically for a university setting. A 1995 analysis into medical school applications by McManus et al found that applicants from BME groups were less likely to be accepted. This was partially explained by them being less well qualified and applying later. However, the study found that having a European surname predicted the applicant’s likelihood of acceptance better than the applicant’s ethnic origin. The researchers concluded that:

“…the process would be fairer if application forms forwarded to universities were anonymous and identified only by arbitrary code numbers, with universities being informed of a candidate's name only for the purposes of interview.”

In recruitment

While there has been little research into the effect of name-blind applications in universities, there has been considerably more research into the effects of name-blind job recruitment.

Wood et al (2009), whose research was commissioned by the Department for Work and Pensions, found that applicants with typically white British names were more likely to be shortlisted for jobs than those with names associated with BME ethnic backgrounds. Similarly, research by Carlsson and Rooth (2007, 2008) found that jobseekers with a Middle Eastern name were less likely to be shortlisted for jobs in Sweden than those with a Swedish name. In addition to having a foreign sounding name, they also found that those with foreign sounding qualifications were less likely to be shortlisted.

The German Institute for the Study of Labour (IZA), has conducted a great deal of research into name-blind job applications. One study examined research from Sweden, France, the Netherlands and Germany. The findings generally indicated reduced ‘unconscious bias’ in call-back rates when applications were name-blind, especially with regard to gender. However, although most findings indicated reduced bias for BMEs, the French research found reduced call-back rates for migrants and residents of deprived neighbourhoods when applications were name-blind. There are potentially numerous national and cultural reasons for varying results between different countries and so we should be cautious before assuming comparability with the UK. The authors, however, argue:

“Firms cannot hire the most qualified workers and simultaneously increase diversity if minority groups have on average lower education outcomes ... if discrimination in other instances leads to differences in this regard, solving this problem is beyond the scope of anonymous job applications.”

Are university applications different to job applications?

The literature raises some differences between university recruitment and job recruitment. It has been argued that these differences may mean that the job recruitment literature does not fully apply to university applications. These differences include:

  • Job recruitment usually selects for a single – or small number of – vacancies, so applicants are directly competing with each other. In contrast, universities usually select a large quantity of students for each course.

  • Job recruitment usually utilises an interview which normally forms the final selection criteria before appointment. Few universities employ interviews as part of their admissions process.

  • Large employers have large human resources departments who perform administrative, verification and initial applicant vetting duties. They are entirely separate to the individuals who are tasked with selecting shortlisted candidates. Many universities have fully centralised admissions services, merging vetting and shortlisting. In smaller universities, these duties may be carried out by one person.

  • There are contrasting legal considerations which regulate job recruitment and university applications. Students are considered consumers and, although there are some exceptions, the great majority of university admissions are not subject to employment law. Both job recruitment and university applications are, however, subject to the Equality Act (2010).

In essence, little research has been carried out into the effect of name-blind applications specifically to universities. Evidence from job recruitment suggests name-blind applications can have positive effects in reducing ‘unconscious bias’ by recruiters. But there are compelling reasons that suggest job recruitment is different from university admissions. Ultimately, therefore, it is not possible to determine the effect of name-blind admissions which UCAS will introduce in 2017 on BME participation in higher education.

James Dobson is a Researcher at Bright Blue.