How to loosen the grip of independent schools on Oxbridge?

Despite the controversy about David Cameron’s recent polemic, the reason why black and minority ethnic students are under-represented at Oxford and Cambridge is more about power relations between state and independent schools than ethnicity.

Oxford, Bristol and Cambridge are the three worst universities for accepting state school students. The 2015 Social Mobility and Child Poverty Commission State of the Nation report stated that Oxford would need to increase its proportion of state school pupils by 24 per cent and Cambridge by 18 per cent to reach the benchmark set by the Higher Education Statistics Agency (which represents “estimates of what the social mix of entrants to each university would be if the university’s intake was representative of the general intake of those with the required grades entering a UK higher education institution in that year”). But it won’t happen. Independent schools maintain a python-like grip on Oxbridge. Bias is not just about gaming the system or accidental privilege. It is about the dynamics of power and influence between independent schools and Oxbridge colleges, and the timidity of government regulation. If independent schools were to lose their grip on Oxbridge, the incentive for parents to spend a fortune on private education would dissipate.

Using statistics, some of which were obtained under the Freedom of Information Act, I compared admissions of sixth-formers from independent and state schools to Oxbridge. I measured the odds of sixth-formers applying, getting an interview and being accepted. Overall, in 2014, Department for Education figures for Oxbridge acceptance from all A-level student entries show that those from independent schools have a five times greater chance of admission than all state pupils. More shocking, Oxford accepted four times as many private independent students (1,194) as comprehensive school students (299). Why so few comp kids? True, many good ones are put off by social elitism, but I expected that interview and acceptance rates for those who do apply would be the same or a bit higher than for independent school students.

Not so. The stats are scarily unequal. Cambridge interviews most applicants, but Oxford discards half before interview. In 2014, applications to Cambridge came from 6,306 state-maintained pupils (3,186 of whom came from comprehensives), and 2,826 independent school pupils. The odds ratios show that independent school applicants had a 53 per cent higher chance of acceptance than state school students overall and an 85 per cent better chance than comprehensive school students. At Oxford in 2014, independent pupils were 58 per cent more likely to be shortlisted for interview than comprehensive school applicants. Of those called for interview, independent school pupils were 14 per cent more likely to be accepted; and overall, applicants from independent private schools were 46 per cent more likely to be accepted than applicants from comprehensives.

These biased odds on interviews and acceptance at Oxbridge, invisible to the electorate, expose the scale of class dynamics that privilege private schools over others – and act as a root cause of growing inequalities among those who run British society. One possible solution is admissions reform. Oxbridge and the Russell Group could drop GCSE A* grades as the main criterion for interview or offers. Parents who have the financial wherewithal for educational apartheid expect better GCSE grades from a private school. Attendance at a state school will cost equally bright pupils several A*grades.

Such pupils, the evidence shows, do better in their degree when matched with independent school pupils for A-level grades. Yet the GCSE premium makes a huge difference to their chances of being shortlisted or accepted at Oxbridge. Current “contextual adjustment” is a fig leaf, with Oxford simply “flagging” a small but unknown number of pupils who live in a deprived postcode and attend a below-average school. Unadjusted GCSE * grades hugely favour independent school students.

AS-level grades are probably a better criterion for interview and entry. State pupils will have had a year of parity in the more academic environment of the sixth form. Generally, comprehensive schools do not put forward applicants to Oxbridge unless they do especially well at AS level, so it would be worth examining whether AS or GCSE performance is a more appropriate measure.

But how to reform interview bias? Leading private schools have departments focused on Oxbridge interview technique to help pupils join what the commission report calls “the members’ club…seen in university interview questions that some are coached to answer but are a mystery to the rest”. In Brent, my London borough, bright comprehensive school pupils are offered a session on Oxbridge interviews at a charge of £20. No surprise that independent school pupils are much more likely to get through the interview. And what of the old school tie? Have admission decisions been analysed by the tutor’s own educational background or when interviewers are blind to the pupil’s school?

Bluntly, then, admissions reform is unlikely to work. The only really effective strategy would be quotas. Oxbridge should admit students in direct proportion to the numbers applying from three strata: independent private, independent state (grammars and academies) and state comprehensives. Quotas would not cut standards because Oxbridge would still find the best. Stratified quotas would incentivise middle-class families to move children from the private to the state sector, relieve them of the huge burden of school fees and in the process improve state education. Such a policy would be egalitarian, cost-effective and dramatically change British class elitism. That’s why it won’t be adopted.

Anthony Costello is director of the Department of Maternal, Child and Adolescent Health at the World Health Organization and an honorary professor in the Institute for Global Health at University College London, although he is writing here in a personal capacity.