said there was a wider dividend too: ‘We have a happier workforce, knowing that the salary system is something they can trust.’22
The new focus on gender is adding an urgency to questions being asked in many different settings including at two of the world’s top universities. Both Oxford and Cambridge have been puzzling over a gender gap at the highest levels of attainment – first-class degrees in some subjects. In 2014, Cambridge’s results in history showed that in the first part of the degree course, 88 per cent of the firsts went to male students, despite there being near equal numbers of men and women enrolled.23 At Oxford that year, 35 per cent of men but only 21 per cent of women studying English gained a first-class degree and there has been a persistent gender attainment gap in chemistry, too.24
At both universities, all students would have entered with excellent exam grades, and the effort to understand the gaps is made more complex by the range of subjects involved: from the essay-based humanities, where marking is more subjective, to the exactitude required in the sciences. At Oxford, Pro-Vice-Chancellor and Advocate for Diversity Rebecca Surender told me they have been looking at everything from pre-university education to the admissions process, exposure to female role models and the style of teaching, for example the often intense interaction in weekly tutorials. ‘The kind of degree you get matters and we don’t want women to be disadvantaged when they leave us and go into the world,’ she says. ‘Preliminary results suggest that there is no single explanation but rather a set of interactions between wider socialisation and what happens before women get to us, together with some environmental factors in the institution.’
At Cambridge one study investigated the relationship between exam structure and performance. Academics in the physics department set up a mock exam for first-year undergraduates in which, for some questions, the usual format was replaced with a ‘scaffolded’ version, broken down into sections showing the marks available for each.25 This style is closer to what those undergraduates would have been accustomed to in their school-leaving exams, and while it resulted in improved performance for all candidates, the women benefited more than the men. On average their marks increased by more than 13 per cent compared with their previous exam performance – while for the men the average increase was 9 per cent.
Dame Athene Donald, one of Cambridge’s most senior female professors and a physicist herself, told me a close focus on the beginning of the university experience was vital: ‘If women come here and struggle in their first year, they may never gain the confidence to proceed. In a subject like physics, where the percentage of girls is only 20–25, and you will be conscious at some level of being in a minority, it can feel even more threatening. Sometimes young women don’t like to say “I’m struggling” because they think that’s an admission of failure, so they struggle in silence.’
She wonders about the impact of stereotypes – perhaps young women don’t expect to do well in a mathematics-heavy subject such as physics – but also why more progress hasn’t been made since her own days as an undergraduate at the university. ‘In my final year class there were eight women out of about 100. But one didn’t expect anything else. I knew perfectly well that there were only three colleges that could admit women. What I find shocking today is that despite all the changes, despite the fact that we are fully co-educational apart from three women’s colleges, we still have these issues.’
At Oxford, one of the studies overseen by Rebecca Surender has focused on academic self-concept, or the belief in your ability to succeed in a particular subject area. Given that this tends to correlate with academic achievement, the aim was to establish any differences between men and women on arrival at the university and how their academic self-concept might change during the first year of study. She told me the findings showed that from the beginning of their course, male students had a higher perception of their own competence in their subject compared with their female peers, but for both sexes the level remained stable over the course of the academic year.
For the university this is of course a welcome finding because it suggests that the difference exists before students arrive. I do wonder, however, if there might also be an impact from the history, tradition and sense of excellence honed over centuries that surrounds you in these environments. I certainly remember times when I found Cambridge intimidating. The Times journalist Sathnam Sanghera, who was born into a working-class Sikh family and went on to read English at Cambridge, remembers ‘negative feelings of unbelonging’ while he was there.26 If academic self-concept is the key, how much might it be affected in people who have a nagging sense that they don’t quite fit or fully deserve to be in such a celebrated place?
In one study, researchers used a different university environment to investigate how peer perception and gender might influence students’ assessments of each other. Sarah Eddy and Daniel Grunspan asked biology undergraduates to complete surveys, getting them to highlight peers they felt were particularly strong in their grasp of the material studied and those they thought would do well on the course.
The results showed that male students were much more likely to rate other men as knowledgeable, a tendency that lasted throughout the academic year and persisted even after controlling for class performance and outspokenness. Female students showed no gender bias, nominating both fellow female and male students. The authors also found that there were some students who stood out in the eyes of their peers and were nominated multiple times, and that these students were always male. It wasn’t as though there weren’t women with similarly high grades in their classes, who also spoke up frequently and demonstrated their knowledge, but somehow they never gained the same ‘celebrity status’ as their male counterparts.27
When I read this study I started to imagine what the scene in the classroom might have looked like. The ‘celebrity’ students would no doubt have been aware of attracting their peers’ attention when they spoke – heads would have turned to listen to them, perhaps nodding in agreement. It’s a good experience to have, one that’s certain to make you feel more at ease, happier with your command of the subject material and probably spurred on to make further points. Could these apparently small interactions build up and develop capacity in a subject so much so that attainment is higher – or the chances of further study or a career in the field are increased? I started to think more and more about expectations of behaviour – whether in a classroom, smaller tutorial-style gatherings of students, or the first day in a new job. If we grow up with assumptions, even ones of which we are barely conscious, that men will speak first or take the lead, that can easily turn into a pattern that validates and reinforces those assumptions.
Consider this alongside evidence on discrimination within the workplace or before people even get there. One meta-analysis of studies conducted in OECD countries over a twenty-five-year period found that discrimination against ethnic minority applicants in the hiring process was commonplace.28 In 2009, research commissioned by the UK government reported high levels of name-based discrimination when researchers sent out multiple applications for real-life openings. The main difference between the applications was the likely ethnicity associated with their name: ‘Andrew Clarke’ was used to denote a white British male; ‘Mariam Namagembe’ for a black African female and ‘Nazia Mahmood’ for a Pakistani or Bangladeshi female. White names were favoured over equivalent applications from ethnic minority candidates.29
More recently, big data analayses have been used to look at information about individuals in new ways. The US-based neuroscientist and artificial intelligence expert Vivienne Ming took a vast data set of millions of real-life professional profiles collected by a tech recruitment firm and used them to compare the career trajectories of software developers whose first names were either ‘Joe’ or ‘Jose’. She found that those named ‘Jose’ typically needed a Master’s degree or more to be equally likely to get a promotion as a ‘Joe’ who had no degree. She called this a ‘tax on being different’, because of the extra costs and time involved in gaining the higher qualifications.
When Ming then used her model to compare the profiles of software engineers with male names against those with female names, she also found a ‘tax’. Typically, women needed a Master’s degree in order to compete with a man with a Bachelor’s degree. No wonder people who face these extra hurdles sometimes decide it’s not worth pursuing a particular path, she concluded: ‘The tax comes