A record number of students have achieved top grades in their A-levels this year. But the results may hide inequalities between students and across different school types. They will also affect university admissions and student numbers in both this and future academic years.
A-level exam results released on Tuesday in England, Northern Ireland and Wales show a record number of students have gained the highest grades. Just under 45% of entries were awarded A or A* grades, up from 38.5% last year. But these results mask inequality and create uncertainty around university numbers.
Figure 1: Percentage of entries gaining ‘top’ A-level grades
Source: Joint Council for Qualifications
In 2020, A-level examinations were cancelled, and replaced, initially with a combination of centre assessed grades (assessed by the student’s school or college) and an algorithm to standardise the grades awarded. But the algorithm led to inequality between students and unfair outcomes for some, particularly those in larger schools and centres without a history of high-attaining students (Proud, 2020). Following public outcry, the controversial algorithm was scrapped, and students were awarded grades based on the best outcome from either the centre assessed grades, or the algorithm result.
At the beginning of the 2020/21 academic year, the announced plan was to return to A-level examinations to ensure that students received fair results. But in February 2021, Education Minister Gavin Williamson announced that, for a second year in a row, they would be cancelled and this time replaced by teacher assessment, which would be moderated by examination boards.
Teacher assessed grades were determined based on work already undertaken and new assessments, either written by the teachers or taken from existing exam-board questions. They also allow for adjustments to take into account individual students’ circumstances, such as if they lost significant periods of learning due to Covid-19 infection (Ofqual, 2021).
Yet this approach raises concerns that students may be treated unequally, both within a particular school and across different types of schools.
Figure 2: Percentage gaining A or A* by school type
Source: Ofqual, 2021
Since 2019, there have been significant increases in the proportion of students gaining the highest grades. This has occurred across all school types (see Figure 1). But the largest increase has been in the independent sector, with a 26 percentage point increase in the proportion of students gaining A or A*.
In addition, within schools, the use of teacher assessment may be problematic because of stereotyping or implicit bias. Evidence suggests that particular ethnic groups may be under-assessed compared with white peers in teacher assessments. For example, at age 11, when key stage 2 scores were decided through teacher-assessment, Black Caribbean and Black African pupils were found to be under-assessed, whilst Indian, Chinese, mixed white and Asian students were over-assessed compared with white peers (Burgess and Greaves, 2013).
Similarly, when grading criteria are poorly defined, studies indicate that assignments signalled as being written by a black author may be systematically under-graded compared with assignments signalled as having a white author (Quinn, 2020). Yet, analysis from Ofqual suggests that, whilst there was evidence of a black/white achievement gap, this has remained consistent over recent years, and the use of teacher assessment in 2021 does not appear to have increased this gap (Ofqual, 2021).
The effects of the use of teacher assessments on different socioeconomic groups have also been cause for concern. For example, on average, UCAS predictions (which are determined by a similar process to teacher assessed grades) over-predict the performance of students, but high attaining students from disadvantaged backgrounds are much more likely to receive predicted grades lower than those that they actually achieved (Murphy and Wyness, 2020). This further raises the risks associated with teacher assessed grades, as high attaining students may have received lower grades than they deserved.
Why have grades increased so much?
There are multiple reasons why the move to teacher assessments in 2020 and 2021 has resulted in large increases in grades. In effect, the assessments of the past two years have been more like coursework than ‘traditional’ exams and there is evidence that coursework is associated with lower anxiety and higher grades (Richardson, 2015).
Further, in high-stakes exams, it is likely that attainment is determined mostly by the ability and preparation of the student, but partly by random events or shocks. These could be from a number of sources, such as the questions that are asked in a particular exam and whether they align well (or poorly) with the topics the student had revised, individual nerves, or even an external event, such as a dog barking nearby (Kane et al, 2006).
The use of teacher assessments has allowed for a greater number of assessments to be considered, which means that the negative effects of these random events can be cancelled out by performance elsewhere.
The incentives facing schools and teachers may also have played a role in the increase in grades. School performance is often an important factor in parents’ decisions about which school to choose for their children. As a result, there may be an incentive for schools to assess a student’s performance based on their best overall achievement during in-class assessments, rather than a more holistic view.
For example, one study shows that, at GCSE level, schools focussed resources on students who were most likely to improve key metrics in league tables, rather than necessarily improving the school’s performance as a whole (Wilson et al, 2006).
Through these mechanisms, the increase in student achievement, particularly at the upper end of the distribution, is not surprising.
How have universities reacted?
Historically, universities make many more offers than they have places available for a course because they are making offers under uncertainty. Firstly, because students apply to multiple institutions, it is unclear what proportion will firmly accept the offer of a place.
Secondly, as universities are contractually obliged to admit students who meet the terms of their offer, if a university has 100 places available on a programme, they might make an offer that they expect 70 or 80 students to achieve. This allows the university to fill up the remaining places either by giving places to students holding offers who didn’t achieve the terms of their offer, or through the clearing system.
But this means that if there is a sudden, unexpected, increase in the proportion of students gaining high grades – as has happened this year – then rather than 70 or 80 students achieving the terms of their offer, there might be 150 students who do so.
At the beginning of the 2020/21 admissions cycle, A-levels were expected to be administered through the normal process of examinations, with a return to normal grade distributions. Consequently, many institutions will have designed their strategy of making offers based on (relatively normal) conversion rates.
The cancellation of formal A-level examinations and the move to teacher-assessed grades led some universities to realise that grade inflation was once again likely. Some institutions, such as Oxford, reduced their overall number of offers in 2021 to account for both the possible grade inflation and the larger than normal 2020 cohort (SchoolsWeek, 2021).
Where universities have not reacted and reduced offers, we may expect to see increased cohort sizes, or incentives introduced to defer study (Guardian, 2021).
Where can I find out more?
- Should students apply to university after they’ve got their exam results? A piece on the Observatory looking at alternative ways for students to progress to university. The government has recently consulted on a system of post-qualification university admissions, where school students would receive and accept offers only once they have their A-level or equivalent grades.
- How should we assess school students now that exams have been cancelled? Observatory article outlining the case for A-level students to receive in-class testing, with flexible timing and content, to take account of differences in their learning experiences.
- How should the academic achievements of school students be assessed this year? A piece on the Observatory highlighting alternative methods of assessment to consider the full context of students’ educational experience.
- Should we stop using predicted A-level grades in university applications? Observatory article on the benefits and drawbacks of using predicted grades for university offers.
