Over the last two weeks, students across the UK have received their GCSE and A-level grades. This year’s cohort were the first to take new “reformed” exams across a range of subjects.
The changes were – like most of the tinkering with our education system – designed to correct perceived failings in how we teach and test our children. Yet there is early evidence that some of them may have tipped the scales based on attributes that have little to do with innate aptitude or ability to learn, and more to do with how we socialise girls to to behave.
In the new GCSEs, girls continued the recent trend and retained their lead over boys, doing particularly well at gaining the highest “nine” grade in the new scoring system. But in the higher grades at A-level, girls fell behind boys for the first time in 17 years.
Across all A-level subjects and throughout the UK this year, boys moved ahead of girls at the A-A* level grades, with 26.6 per cent achieving one of the two top results. That fits with the long-term pattern which has seen boys slowly closing the gap with girls. But their big jump from 0.3 per cent below girls in 2016 to 0.5 per cent ahead this year, is an anomalously large leap forward – why?
There are many possible reasons. More boys may have taken more subjects which score well. Or more boys who would not have done as well as their peers in the A-level structure could have opted out of A-levels in favour of vocational courses. But there is one major change to the A-level exam system that could have had a huge impact – the shift towards less coursework and less modular testing.
Some academics had already predicted that such change would work against girls. Research from Professor Alan Smithers from the University of Buckingham has pointed out how girls’ results started to climb with the introduction of modular exams in 2002. Meanwhile, Spanish researchers have shown that girls perform best in classroom tests rather than in national exams, says Professor Therese Hopfenbeck, director of the Oxford University Centre for Educational Assessment.
These conclusions appear to be supported by the results from the new A-level exams. Within England’s 13 new reformed subjects, 23.4 per cent of both genders gained an A or A*. But the proportion of all girls in this group fell by 1.1 percentage points from 2016’s results, compared to just a 0.2 point fall for boys.
According to Jo Boaler, a former teacher in London and now a Professor of Mathematics Education at Stanford University, research shows girls and boys are equally capable across various systems of testing, but girls are socialised to be more anxious in some test conditions. “In the PISA results which came out recently we saw boys ahead of girls in maths. But when they factored in levels of anxiety those differences completely disappeared,” she says. Similarly, coursework is a less anxiety-inducing method of testing, where you can show what you can do in a relaxed environment.
It is also likely that high-achieving girls are most susceptible to feeling anxious about their own abilities. “One of the things we know about mindsets is that the people with the most fixed mindsets across the whole school system are high achieving girls,” explains Professor Boaler. “These are the girls that we particularly need to think positively about themselves and believe in themselves. And changing from coursework to an exam system would definitely have an impact on that.”
This is something I wish I’d known a decade ago, when I received my own AS-level exam results. Once again I’d under-performed in my most loved subject, English, and it was only the knowledge that I was good at coursework (plus the encouragement of a brilliant new teacher freshly arrived at my Devon comprehensive from Westminster boys’ school), which gave me the courage to apply for Oxford anyway. Three years later I’d finally gain enough confidence in my exam technique to get a first class degree.
So why did the government move away from the coursework and modular methods of testing? In 2014 Ofqual chief executive Glenys Stacey said that “non-exam assessments do not always test the skills they are meant to assess, they can disrupt classroom time better spent on teaching and learning and may provide limited evidence of performance across a group of students if they all get limited marks.”
But Professor Boaler is concerned that the current government has not properly considered the impact that moving back to more linear, timed testing has. In the USA, schools and districts are pulling back on maths fact tests (speed testing on times-tables), she says, which are now thought to be responsible for the onset of maths anxiety in kids. In contrast, the UK has automated them so that students are now constantly being tested by computers. “On the question of why aren’t the government doing things differently, I don’t even know what to say other than that the research is there and they don’t care about it.”
Professor Therese Hopfenbeck is similarly concerned that a lack of consideration has been given to how exams are experienced across social divides, not just those of gender: “It’s harder for students who are coming from a poor background to be able to succeed and believe in themselves than students from a higher economic background – and this is something we see in all the previous research studies. I think that is of concern for our democracy.”
The result is a worrying sense that in a rush to appear “tough” on educational standards, the government has risked hurting the ability of students to meet their full potential when it matters most.