This year’s outcomes explained after teacher-assessed grades
Summer 2021 arrangements
This year AS and A level exams did not take place because of the disruption to students’ education caused by the coronavirus (COVID-19) pandemic. It would not have been fair for exams to take place in the usual way. Instead, grades were determined by teachers based on a range of evidence. These were known as teacher assessed grades, or TAGs. Since the pandemic caused different levels of disruption across the country, many schools and colleges had not been able to teach all the course content. In response, students were only assessed on the content they had been taught while ensuring sufficient coverage of the curriculum to enable progression. Centres were given flexibility to decide how to assess their students’ performance, for example, through mock exams, class tests, and non-exam assessment already completed. This flexibility was to accommodate ongoing disruption to teaching including school closures and the need for students and teachers to self-isolate.
Schools and colleges put in place internal quality assurance processes. They were required to make sure at least 2 people were involved in each judgement and the head of each school or college had to sign off the grades. Teachers and senior leaders in schools and colleges have worked hard to make sure that judgements were made, quality assured and submitted on time. They also submitted examples of students’ work to the exam boards.
Exam boards put in place external quality assurance arrangements, checking each centre’s policy, reviewing the profile of grades submitted, and reviewing samples of student work. Where exam boards had concerns, these were followed up with the school or college and in some cases, teachers reconsidered their judgements and submitted revised grades.
These assessment arrangements are quite unlike those typically used. Normally students sit the same assessments, often exams, at the same time and under the same conditions, and they are marked and graded in the same way. Without the mechanisms usually used to secure standards over time and between students, outcomes this year look different from previous years – as we expected they would.
A level students are generally more able to study independently than younger students and have more study time for each subject. There has been an increase in outcomes at the top grades and stability at lower grades. This may be because the most academically able students were most able to study independently.
There are some small changes in the gaps between the average results of some groups of students, which are observed in normal years, which may reflect the impact of the pandemic on learning . It’s likely, however, that these changes would have been greater had we had not given teachers the flexibility to assess students only on the content they had taught.
- Overall A level results are higher at grade A and above compared to 2020 (44.3% in 2021 compared with 38.1% in 2020, and 25.2% in 2019).
- Similar patterns in outcomes at grade A and above are seen in Wales (48.3% in 2021 compared with 41.8% in 2020, and 27.0% in 2019) and Northern Ireland (50.8% in 2021 compared with 43.3% in 2020, and 30.9% in 2019).
- Results are also higher at grade B and above compared to 2020 (69.8% in 2021 compared with 65.4% in 2020, and 51.1% in 2019), whereas results are relatively stable at lower grades (C to E).
- The overall increase in outcomes this year compared to previous years may partly reflect the change in assessment arrangements, for example, teachers may have given students some benefit of the doubt across the multiple opportunities many students had to show what they had learned – quite different from end of course exams.
- Analyses show general stability in the differences in outcomes for students with different protected characteristics compared to previous years and increases in outcomes for many groups. This suggests that the changes to the assessment arrangements have lessened the unevenness in outcomes we may otherwise have seen.
- Students who think there has been a mistake with their grade can choose to appeal via their school or college; students who are unhappy with their grade can choose to sit an exam in the autumn.
Data for NI in 2020 and 2021 includes Health and Social Care (single and double award)
Summer 2021 results
This summer we have carried out a range of analyses to support understanding of results.
Variability in centres’ results
We compared the level of variation in centres’ A level outcomes in 2021 to that seen in 2020 and 2019. Higher variation means centres’ outcomes have fluctuated between years; lower variation means centres’ outcomes have been stable between years. Variation in centre results was generally lower than it was in 2019 and only slightly greater than it was in 2020. This suggests that the assessment arrangements lessened the impact of the pandemic on centres’ results – otherwise we would expect to see much greater variability in centres’ results this year compared to the past. For example, the variation (standard deviation) in centres’ outcomes for A level maths was 13.6 in 2019, 11.12 in 2020 and 11.21 in 2021.
Exam boards looked at student work from a sample of schools and colleges. They found no evidence that teachers’ judgements were systemically biased in favour of one group of students or another. Nonetheless, we have conducted equalities analyses to consider whether the gaps between the average results of different groups of students, which are observed in normal years, have changed this summer. This has included a consideration of gender, ethnicity, free school meal eligibility (FSM), English as an additional language (EAL), socioeconomic status (SES) and special educational needs and disability (SEND). We compare the extent to which the relationship between results and student background variables in 2019 and 2020 are similar in the 2021 outcomes. In doing so, our model takes account of the prior attainment of the students as well as other factors. The analyses show general stability in the differences in outcomes for students with different protected characteristics compared to 2019 and 2020 and increases in outcomes for many groups. There are some, small, relative changes in overall outcomes. Between 2019 and 2021 the gap between male and female students has reversed and has grown compared to 2020, with females receiving higher grades on average than males. The average change is a fifth of a grade. Our model suggests that in 2019, SEND students received slightly higher grades than non-SEND students. In 2021, this has reversed with non-SEND students receiving slightly higher grades on average than SEND students. The change is a tenth of a grade. In the raw data, SEND students have historically performed lower than non-SEND students. It seems likely that many of these changes reflect the uneven impact of the pandemic and that the changes to the assessment arrangements may have lessened the unevenness in outcomes we may otherwise have seen. The details of our analyses are included in our technical report.
Centre type results
We have compared the results of different types of centre across 2021, 2020 and 2019. Outcomes for all types of centre have increased compared to 2020. The extent to which they have increased may partly reflect longstanding differences in the distribution of grades for different centre types. For example, in normal years, the distribution of grades for students attending independent centres is clustered around the top grades. A uniform increase in grades awarded across all types of school and college is therefore most likely to benefit students attending independent schools at the top grades. For other centre types, the impact of increased grades may be lower down the grade distribution – wherever most of their candidates are normally clustered. This clustering of students at different points in the grade distribution means it is helpful to consider both the relative and absolute changes in outcomes across centre types. This is because relative changes on small numbers often look big but relative changes on big numbers often look small. On the other hand, absolute changes on small numbers often look small but absolute changes on big numbers often look big. Therefore, it is important to explore both types of change.At grade A and above, the largest relative increase is for ‘other’ centres (+22.5%), further education establishments (+21.4%) and secondary comprehensives (+18.7%), and the smallest increase is for sixth form colleges (+12.0%). The increases for independent centres and academies at grade A and above are partway between the highest and lowest changes (+15.2% and +15.8%, respectively). At grade C and above, the relative changes are similar across centre types, ranging from -0.1% to +2.2% compared to 2020. The smaller changes at grade C reflect the overall stability in outcomes at this grade compared to 2020.In absolute terms, the increase in outcomes at grade A and above is largest for ‘other’ centres (+9.4 percentage points (pp)) and independent centres (+9.3pp). At grade C and above, outcomes are relatively stable for all centre types, reflecting the stability in overall outcomes. The largest increases at grade C are for ‘other’ centres (+1.9pp) and free schools (+1.6pp). Exam boards looked at student work from all types of school and college as part of the external quality assurance process. They did not find that any type of school or college was more likely than others to have provided grades that did not reflect the standard of their students’ work. Indeed, they found that, irrespective of the type of school or college, the grades were largely supported by the quality of students’ work. It is also worth noting that more able students are also generally more capable of independent study.
JCQ have published regional results compared to 2020. The rank order of regions according to their results is broadly similar to 2020 and the percentage of students achieving grade A or above has increased in all regions. As is the case every year, there is some variation in the size of the increase across regions. At grade A and above, the greatest increase was seen in London where the rate increased by 7.2pp compared to 2020. At grade C and above the greatest increase was seen in the East Midlands where the rate increased by 1.6pp compared to 2020. As with the differences seen across centre types, these regional changes will partly reflect the well-established clustering of students around different parts of the grade distribution.
Today (10 August 2021) we are publishing:
Published 10 August 2021