“As I was saying when we were so rudely interrupted…”

These words, so the story goes, were the first uttered on BBC television when transmissions resumed after the second world war. And today, 1,099 days since we last experienced a ‘normal’ A level results day, we too have a moment of resumption.  Sadly, that BBC story is just a triumph of apocryphal storytelling, and the actual words Jasmine Bligh spoke on that day were ‘Good afternoon, everybody. How are you? Do you remember me?’ - which to be fair does have its own charm. Today has elements of both the myth and the reality of that moment. COVID was (to say the least) a rude interruption, and a great disruptor of our normal routines and practices. There is a chance also that we don’t ‘remember’ results day, and need to re-learn the appropriate responses to what we find. 

It is of course too early to have gathered and processed the data required for meaningful and nuanced analysis, but we can identify some headline features in the data, consider the implications of these features, and set an agenda for the things we need to explore further. 

These, then, are my early thoughts on the results:

1. The results are in, and national grade profiles are high. A* grades are close to 15%, and 36% of students secured A* or A grades. Back in normal times (2019) the figure for A*/A was 25.5%. Immediately we see that we will have to recalibrate our expectations regarding what ‘good’ results look like. What will take longer to establish is what these results really mean: how these results have been distributed across sectors, institutions and subjects, and equality and diversity monitoring groups. Sixth form colleges will largely be delighted with this year’s outcomes, as we were (typically) very modest in our 2021 awarding, and the level of generosity baked-in this year is higher than that generated overall in sixth form colleges in 2021. 
2. We can now get a sense of ‘how high’ grades are: We can generate a very simple measure of how generous results are by looking at how the average grade shifts from year to year. Back in 2019 (using a scale where A*=6, A=5, B=4, C=3, etc.) the average A level grade was 3.49; in 2022 it was 3.88. This increase of 0.39 suggests that 39% of grades were higher than those in the 2019 base year (as 6.8% of grades move from an A to A*, 10.9% of grades move up from B to A, 11.2% of grades move from C to B, and so forth). Given the generosity of TAGs in 2021, it is not surprising that the AQA pre-results day message suggested that it was unlikely that any schools or colleges would have results higher than in 2021. What they have not factored in is the relative caution (some would say integrity) with which sixth form colleges generated teacher assessed grades last year. I suspect that our sector will see many colleges with, in raw terms, genuinely record results.
3. The expected 50:50 weighting of 2019 and 2021 results has materialised but there is a slight bias towards 2021 standards Using our average grade measure, in 2019 the average grade was 3.49, and in 2021 it was 4.19, a gap of 0.70. The 2022 figure of 3.88 is higher than the middle value between 2019 and 2021 (which would be 3.84).
4. The level of generosity varies significantly from subject to subject One curious decision that Ofqual made was to not set an ‘overall’ level of generosity for all subjects, but to set standards relative to how generous each individual subject was in 2021. This is great news if you are a drama student (where 90% of students received an inflated grade in 2021), but less good news for mathematicians, where only 59% of grades were inflated. It is here that the most important ‘story’ of results day lies: while overall grades are 39% up on 2019, in Drama grades are 64% up and in Music it is 72%, while in Chemistry (32%), and Maths (26%) grading has been far less generous (and is closer to 2019 than 2021). Your chance of getting an A* or A in Maths is 7% higher than 2019, and in Music it is 23% higher. This is not terribly helpful for university entrance either, where the subjects with the most inaccurate teacher assessed grades in 2021 give the best 2022 grades in the scramble for places
5. Any model of analysis which uses historical benchmarks will be largely worthless. Anyone trying to use historic L3VA to assess value added, or indeed any model which relies on a historical benchmark, risks arriving at misleading judgements and implementing misguided actions for improvement. We already knew that the prior attainment profiles of this cohort of students are very different to those in previous years. What we have learned today, though, is that the level of generosity in the final grades is historically unique and varies widely from subject to subject. 
6. We risk deluding ourselves through the use of historical data My advice would be (except where it looks like something has gone seriously wrong) to wait until you have a full value-added analysis based on benchmarks derived from 2022 outcomes before passing judgement on quality or attempting to build a picture of performance for this year. It will be a rare college indeed which has not produced results above those attained in 2019, and many results-day press releases will (entirely reasonably) claim record results. Ultimately, though, to quality assure properly we need to know how well our students performed compared to similarly-qualified students taking those subjects in other colleges. It is the progress they made, not the standards they attained, that is important. On a different note here, if standards do return to pre-pandemic standards next year, those subjects which have seen the highest generosity will need to prepare themselves for the biggest drop, and build this in to their monitoring of student progress.
7. We have a pressing need for intelligent balanced analysis of how grades landed in individual subjects, and with different groups of students. The essence of the Six Dimensions work over the coming weeks will be to develop a picture of how grades played out across the sector. By developing a picture of what is normal, we can help colleges identify which of their subjects performed genuinely well, and those where underperformance needs exploring. There will be high level insights as well – performance across different equality and diversity monitoring groups compared to normal years, and building an understanding of the relationships between 2019 results, CAGs, TAGs and 2022 outcomes.
8. We will never know what difference the advance information and other adjustments made as the grade profiles were set statistically: while examiner judgement did have a role to play this year, the grades available to students were set by computer. If advanced information helped all students, some students or no students, this will have had little or no bearing on results. Quite how it was expected that eighteen-year-olds could navigate advance information without introducing new unfairnesses to the process is a bit of a mystery to me. Our analysis will be able to shed some light on this at group level. We will be able to identify if some groups have struggled more than others, but will never know the balance between pandemic factors and assessment factors.
9. Ultimately what these results represent is a return to results which reflect how effectively students were served in different institutions, rather than how generous each institution was prepared to be in the awarding of CAGs and TAGs. It is three years since we have had the chance to have external validation of how well our subject departments are performing. While we need to be more cautious than we might have been in pre-pandemic times, it is imperative that we learn from the results, and build our understanding of how best to meet the needs of our current and incoming students.

Over the coming week, we will be gathering data from as many colleges as we can with a view to be in a position to start sharing more considered analysis as close to the start of term as possible. The need for intelligent, balanced and nuanced analysis is as high today as it has ever been: we do after all need a powerful antidote to a media which seems to find new ways to undermine students each year. We owe it to them to analyse their performance carefully, and use what we find to support our current and future students.