Ofqual’s guide to the 2020 AS and A level results in England.

Key points

  1. In the context of exams this summer being cancelled, we had 2 broad aims: to make sure as many students as possible could get grades so they could move on to the next stage of their lives and to do that in as fair a way as possible.
  2. Overall A level results in England are higher at grade A and above compared to 2019 (27.6% in 2020 compared with 25.2% in 2019). This reflects the approach to awarding grades this summer and the decisions that, where possible, have been taken in students’ favour, as part of the standardisation process.
  3. The majority of grades awarded to students are the same or within one grade of the centre assessment grades (CAGs) – 96.4% at A level and 91.5% at AS.
  4. Students who are not happy with their calculated grade can ask their school or college about the appeals system or they can choose to sit an exam in the Autumn.
  5. Entries for reformed AS qualifications in England have dropped by over 26% compared to last summer. This makes it much more difficult to interpret any changes in year-on-year results.

Today (13 August 2020) we are publishing:

You may also find it useful to read about how we regulate GCSEs, AS and A levels in England.

Summer 2020 arrangements

Following the closure of schools and colleges to most students in March, and the cancellation of summer exams, we have worked with others from across the sector to develop the fairest possible approach to awarding grades in GCSE, AS and A levels, in the absence of any exams. This is to allow as many students as possible to move on to the next stages of their lives, without being further disrupted by coronavirus (COVID-19). We have consulted widely with students, parents/carers, equalities groups, teacher associations and exam boards on our proposals and there was broad support for the approach we set out in April.

We asked schools and colleges to submit centre assessment grades (CAGs) – the grade the student would have been most likely to achieve if they had sat their exams – and a rank order for each subject. There was no opportunity to develop a common approach to grading across the many thousands of schools and colleges. Therefore, we developed a statistical standardisation process so that there was a level playing field for students, regardless of their school or college.

In general, the CAGs submitted were optimistic. This is understandable and in line with the evidence from previous research. Our recent interviews with teachers who’ve been through the process this summer confirms that – they told us that they tended to think about how each student would perform on a good day, while knowing that every year some students have bad days. This was particularly the case for borderline students.

The combined effect of this optimism, if CAGs had been accepted, would have been an unprecedented increase in overall outcomes. For example, at A level, the CAGs at grade A (and above) were 12.5% higher than outcomes in 2019. This would far exceed any overall variation seen in a typical year and would undermine the credibility of students’ grades. Accepting CAGs would also mean that any leniency or severity in the CAGs submitted by individual schools and colleges would not be addressed. This would make it easier to get a grade at one school or college than another leaving unfairness between schools and colleges. In developing the standardisation model, we consulted, and received broad support for a model that aimed to:

  • provide students with the grades they would mostly have achieved if they had sat exams
  • was common across exam boards and subjects for as many students as possible
  • avoided any systematic advantage or disadvantage to particular student on the basis of particular protected characteristics or socioeconomic status
  • was transparent
  • was deliverable in a way that could be overseen by Ofqual

We have published details of the standardisation model in a technical report. In developing the model we have sought, where possible, to make decisions that work in students’ favour when awarding grades this summer. The technical report also provides information on the approach to awarding grades to centres with small cohorts, where it would not be defensible to rely on the statistical evidence.

As in other years, we have used statistical predictions at the cohort level to guide overall national outcomes. These predictions use the relationship between students’ prior attainment and results in a reference year, and use this relationship to predict the expected outcomes in the current year. In the absence of any evidence of student performance this year, the predictions are key in ensuring that, as far as possible, overall standards are maintained, as the Secretary of State requested in his direction to Ofqual.


We have conducted equalities analyses to consider whether any demographic or socio-economic groups of students have been advantaged or disadvantaged by the process of awarding grades this summer. This has included a consideration of gender, ethnicity, free school meal eligibility (FSM), English as an additional language (EAL) students, socioeconomic status (SES) and special educational needs (SEN). We considered the extent to which the relationship between results and student background variables in 2018 and 2019 would be maintained in the 2020 outcomes.

The analyses conducted shows no evidence that this year’s process of awarding grades has introduced any systemic bias. Changes in overall outcomes for students with different protected characteristics and from different socio-economic backgrounds are similar to changes seen between 2018 and 2019. The details of our analyses are included in our technical report.

Summer 2020 results

Overall, A level results at grade A and above are higher than in 2019, by 2.4%. This is not surprising given the approach to awarding grades this summer. As in any year, there is variation in outcomes by subject. There is greater variation in some subjects than in a typical year given the approach to awarding grades, in particular the approach to awarding grades to centres with small cohorts. For every school and college, we have used the most reliable available evidence and where the number of students is much smaller, the statistical model is less reliable. In those cases, we have therefore relied more on the CAGs. As a result, in subjects where there are more small cohorts (classes of 15 students or fewer), we have seen larger year-on-year changes.

Our analyses show that the majority of the grades awarded to students are the same or within one grade of the centre estimates – 96.4% at A level and 91.5% at AS – reflecting the care and professionalism with which schools and colleges have approached the task.

Number of A level qualifications per student

JCQ published data presents the numbers of entries and certifications, rather than data at student level. This is because students typically take AS and A levels with more than one exam board. It is also worth noting that many students take AS or A levels alongside other qualifications, which we have not included in this analysis.

We have combined the exam board data to look at the average number of A levels per student for 18-year-olds in England entering at least one A level each year (students generally complete their A levels at age 18). This is shown in the table below. For A level, the average number has remained stable since 2016 and that continues this year.

Average number of A level qualifications per student (18-year-olds in England)

2016 2017 2018 2019 2020
A level 2.68 2.67 2.68 2.66 2.67

We have also considered the number of A levels that each individual student was entered for. The table below shows the percentage of students that entered one A level, 2 A levels and so on. In recent years, there has been a trend towards a higher percentage of 18-year-old A level students taking 3 A levels.

Percentage of students by the number of A levels taken per student (18-year-olds in England)

A levels 2016 2017 2018 2019 2020
1 12.1% 11.3% 10.5% 10.6% 10.5%
2 16.2% 16.8% 16.8% 17.0% 16.4%
3 63.6% 65.4% 66.9% 68.0% 68.7%
4 7.6% 6.2% 5.6% 4.3% 4.4%
5+ 0.4% 0.3% 0.2% 0.1% 0.1%

AS qualifications

The reformed AS qualifications in England are standalone qualifications that do not count towards an A level. Students therefore do not have to take the AS qualification if they intend to enter at A level.

Following reform, AS entries have declined, and this has continued this summer. Entries are down from just over 1.1 million in 2016 to 70,500 this year. This means that entries to some AS subjects are now relatively small. Where the cohorts are small and changing year on year, outcomes are likely to change, making any comparisons over time difficult.

Instead, we have analysed the number of 17-year-old students taking at least one AS qualification in 2020 compared to previous years. This is shown in the table below. The number of students taking at least one AS qualification has declined significantly since 2017, following a smaller decline between 2016 and 2017.

Number of students taking at least one AS qualification (17-year-olds in England)

2016 2017 2018 2019 2020
AS 270,500 209,540 64,810 40,880 27,100

Published 13 August 2020