This blog was kindly contributed by Neil Renton, Headteacher of Harrogate Grammar School, Red Kite Learning Trust and an Association of School and College Leaders member.
We have all stood outside the examination hall talking about what we hoped would be in the exam and discussing what we had revised. We remember that sinking feeling as we found our name on a seating plan and walked into the hall. Finally, we sat, usually in the warmth of summer, listening to someone reading regulations before we heard the words ‘you may begin’. But our long-established understanding and assumptions about how student achievement is recognised and rewarded have been shattered by the pandemic. A shift from a final examination, marked by the exam board, to teachers being responsible for making reasonable judgements.
When the Government announced in early January that exams would not be going ahead it brought back unhappy memories for teachers of algorithms, the ranking of students and grade quotas, Government U-turns and the unacceptable way young people were treated in 2020. A consultation followed from Ofqual and by the end of February we had received the outcome. With mounting pressure from students seeking to understand what this meant for them, we set about communicating our plan. There were four defining features of the system.
- Grades were to be based on reasonable teacher judgement.
- Students were to be only assessed on what they had been taught.
- Students were to be told the evidence that was going to be used to reach the grade.
- Assessments were to continue over the coming months.
Students returned to school on 8 March after nine weeks of remote learning. We dedicated teaching time to consolidating learning while giving students the opportunity to take further assessments. We spaced out the assessments and we were clear with students on what aspects of the course they would be assessed on. For both Year 11 and Year 13 their mock examinations had mostly taken place at home, so we needed to ensure that the most recent data was captured under higher levels of control. We achieved this by conducting the assessments under the supervision of teachers and invigilators in their classrooms and latterly with blind marking to ensure greater objectivity.
How schools organised provision during lockdown and the assessment processes will have varied significantly. And that is one area where the pandemic really creates a variable.
All children have had a different experience of the pandemic. In our context, our remote provision developed rapidly during the first lockdown so when the second lockdown came, we were able to deliver a full live timetable as all students have an iPad. Although challenging, students continued to learn and assessments continued to take place.
Secondly, from the perspective of students, while they were studying in Year 10 and Year 12 they did not know that they would not be sitting exams. Some looked back to the evidence they had accrued in early assessments and thought they had demonstrated strong performance, others looked back and felt fear that they had not tried hard enough. Equally, some looked forward and thought ‘I have a chance to improve over final assessments’, others thought ‘how do I maintain my performance from the first year?’ Inevitably, however much we tried to reduce the pressure, in the context of the disruption of a pandemic, students worried – the cancellation of exams did not mean the cancellation of worry.
Thirdly, while students were disrupted from October to January with whole year group closures, we had a positive spell from March to June where there were no Covid cases which allowed us to collect robust data. The key point to stress is that of difference and variation between schools and for students.
We then faced the challenge of writing our centre policy that would dictate how we reached the final judgements for the grades that we would issue. We wrote our centre policy, submitted this to the portal, received a telephone check and then received written confirmation from the Joint Council for Qualifications (JCQ) that our policy had been accepted and that Stages 1 & 2 were complete. At this point, we worked with our local network of schools to moderate work, giving us confidence that we were grading student work correctly and consistently.
By the end of May, all the work had been assessed, students had effectively finished and we had amassed sufficient evidence for two data analysts, working without student names across the timeline of evidence based on an understanding of which assessments were completed under the highest levels of control, to propose a grade. We then issued these proposed grades for the subject teachers to use their reasonable judgement to either approve or give a reason for a higher or lower grade. The next stage was to review the teacher grades again based on the historic pattern of performance and ask whether we had sufficient evidence: were higher or lower grades justified; had we applied special consideration; and had reasonable teacher judgement been applied? Our final step was to check the data at a whole school level, using an external benchmarking tool to see if there were any patterns of severity or generosity.
On 18 June we submitted the grades, uploading over 4,000 grades for 285 students at GCSE and 300 students at A-level. We had reviewed approximately 19,000 data points for assessed work and the reasonable judgement of each teacher, over a period of three weeks, before uploading the final judgements. While indebted to my colleagues who have managed tremendous workloads, we have learnt that it is possible to arrive at judgements based upon a robust process where students do not sit terminal exams.
I think it is important to note that behind every grade that university admissions tutors see in the next few weeks are a varying set of circumstances faced by students and schools. As a school leader, after sending requested samples for only three subjects, I hope that the exam boards will accept all the grades so we do not face the shock of last year, when 40 per cent of our grades were initially downgraded.
This generation are remarkable in the circumstances in which they have learnt and I hope that the transition to the next stage in their education is supported without trauma caused by government intervention, with empathy and understanding for what they have faced.
This is a good description of what you did, but it’d be interesting to know what you observed, too. For example, you say:
“Our final step was to check the data at a whole school level, using an external benchmarking tool to see if there were any patterns of severity or generosity.”
Did the tool find any such patterns? (And: what was the tool, and how does it work?)
More generally, what proportion of grades were changed, and by how much, at each step of the process (“reasonable judgement”, review, external benchmarking)?
This is a brilliant description of what has clearly been a well-designed, rigorous and thoughtful process to determine Teacher Assessed Grades.
Which prompts a question: should we aim to keep at least some of this process in the future? Even if, in 2022 and beyond, a return to assessment through high-stress sessions in a sweaty exam hall becomes possible again, would it not be better to combine that with some of the techniques of teacher assessment set out here?