David Cameron has announced that applicants’ names will be removed from UCAS forms from 2017 in an effort to combat ethnic inequalities in admissions to top universities. The prime minister’s announcement comes in response to evidence that British ethnic minority applicants to highly selective universities are less likely to be offered places than white British applicants.
My own research on admission to Russell Group universities, for example, found that offer rates were between seven and 16 percentage points lower for applicants from British ethnic minority backgrounds than for white British applicants after taking into account their A-level grades and the popularity of the courses applied for. Other studies also suggest that applicants from lower social class backgrounds, disadvantaged neighbourhoods and state schools are less likely to be offered places by top universities than applicants with similar qualifications from more advantaged backgrounds.
Unconscious bias happens by our brains making incredibly quick judgements and assessments of people and situations without us realising. Our biases are influenced by our background, cultural environment and personal experiences. We may not even be aware of these views and opinions, or be aware of their full impact and implications.
Evidence from experiments carried out in the US suggests that unconscious bias can be triggered by names alone. One study showed that identical emails from prospective postgraduate students were more likely to receive a response from US college professors if the sender’s name indicated they were white rather than African American, Hispanic, Indian or Chinese. In another study, identical applications for jobs as laboratory managers in US universities were rated more positively when the candidate had a male name rather than a female one.
To my knowledge, no study has explored whether unconscious bias affects UK university admissions decisions. But given that admissions selectors currently see applicants’ names on UCAS forms, the possibility that unconscious bias creeps into the decision-making process cannot be ruled out.
Risk of unfairness lingers
Removing names from UCAS forms will probably help eliminate ethnic and social disparities in university offer rates for comparably qualified candidates – but not as much as we might hope.
Admissions selectors will still see each applicant’s home address, the school they attended, what they have written about themselves in their personal statement and what their teacher has written about them in their reference. All of this may provide subliminal clues as to an applicant’s ethnic and social background. Where applicants are interviewed as part of the selection process, the scope for unconscious bias becomes wider still.
So simply removing names from UCAS forms will not be enough to safeguard against the risk of unfair admissions decisions. Effective strategies to eliminate the influence of unconscious bias on the admissions decision-making processes are also needed. This will require universities to improve the objectivity, transparency and accountability of their decision-making. They should also ensure that admissions selectors are trained to recognise and resist unconscious bias.
Taken seriously at last
Although name-blind UCAS forms will not eliminate the issue, the proposal is still a welcome development. It signals that the issue of ethnic inequalities in university admissions is finally being taken seriously.
Concerns about ethnic and social inequalities in offer rates from top universities have been dismissed time and again by the Russell Group of top UK universities. They have also been downplayed by UCAS in its own analysis of the data. The prime minister’s announcement indicates that a head-in-the-sand response to ethnic inequalities in university admissions is no longer tenable.
If we want to know whether name-blind UCAS forms reduce ethnic and social inequalities in university admissions, researchers need access to complete and detailed anonymised UCAS data. For the past few years, UCAS has been unwilling to share this kind of data with researchers.
UCAS recently agreed it would begin sharing data from 2017, but only for applicants who actively opt-in to share. If a substantial number of applicants don’t choose to opt-in, researchers will only be able to access data for a distorted sample of applicants and any research results may well be inaccurate.
The upshot is that we cannot reliably assess whether or not name-blind admissions increase the fairness of university admissions until UCAS agrees to share fully representative and detailed non-personal data with researchers.