Exams have been cancelled, and students are still going to be awarded GCSE grades.
Normally, grades would be based on 6 exams being completed by students, and the number of marks they score determines where they are in the distribution (i.e. their rank order), and ultimately their grade.
The grades achieved by students are independent of teacher’s views of the students, and dependent solely on the responses they produce to the questions in the exams. How many marks they score depends on:
- How much knowledge is in their long term memory
- How much, and the quality of, revision is completed throughout Year 11, including in the last two months
- Students’ access to resources to support their exam preparation (e.g. quiet space at home, healthy food, good sleep, emotional wellbeing, support from peers)
- A bit of luck, as the exams assess a sample of the curriculum, which may or may not be the parts they were most well-prepared for
This year the exams aren’t happening. To award grades fairly, we cannot base our grades or rankings on our feelings about the students, because we are human and biased in ways we are and aren’t aware of. A quiet student who we think won’t do well, may have done fantastic revision and scored grades higher than we expected. A bright student may have been overconfident and prepared poorly and scored much lower than we expected.
Looking at last year’s cohort for example, only about a quarter of students achieved exactly the grade they were predicted at AP3. Over three quarters achieved equal to or higher than predicted.
In summary, teachers are OK, but not that good, at predicting grades! Which is generally not a huge problem. Also, teachers know some students better than others.
But that is what we are stuck with this year.
We must accept that the grades awarded will not be particularly valid, and must take pressure off individual teachers so they don’t feel personally responsible for the grades these students achieve. Students will be able to have another go in the autumn.
Therefore we have looked at this cohort’s achievement to date, and the previous cohort’s achievement throughout year 11, to come up with as valid a process as possible.
Centre assessed grades
We were asked to produce “predicted grades” before Easter.
- Y11 took Paper 1 mocks in December, and Paper 2 mocks are done in class near Easter
- About half of the classes did not complete Paper 2 mocks before school closed
- Paper 1 mocks, therefore, represent the most comprehensive GCSE assessment of students’ performance that virtually all students completed
We analysed previous cohort’s Paper 1 mock results in comparison to their final grades
- This was used to produce detailed guidance on how to come up with a prediction based on Paper 1 mock grades, along with knowledge of the students’ performance and work rate since January, including any Paper 2 mock data that we had
- This was used to initially set predicted grades
- Teachers then had the opportunity (through a shared spreadsheet) to amend, with comments, or say they agreed
This produced “predicted grades” which will ultimately become our Centre Assessed Grades (CAGs).
Putting students in a rank order is not something we usually do. When we have large comprehensive assessments it’s easy (simply rank by score in the exams).
We do not want to get students to do any further work to generate grades or rankings, as we cannot control the conditions in which they do this work, and we know there will be huge variability in students electronic access, not to mention the physical environments in which students complete work (all of which is controlled in the classroom or in the exam hall).
We propose to carry out a similar process to that used for generating predicted grades:
- Give teachers another chance to look at the predicted grades, with an opportunity to change any (with justification noted)
- Sort all students (for each of the 4 courses – combined, biology, chemistry and physics) by their total score in the three Paper 1 mocks (or scale their total up if one or two papers were missed)
- Sort by the predicted grade we have assigned
- Use this order to rank the students in each grade
- Re-sort by class
- Share this with the team, giving teachers the opportunity to comment on those in their group (e.g. “should go up” “should be below student X” or “shouldn’t be close to the grade boundary” stating their evidence/justification). Teachers will just be looking at the students in their class at this point.
- We (Head of Science and KS4 lead) will then go through this list, making the necessary adjustments to the overall ranking, based on individual teacher’s comments
- The full spreadsheet is shared with the team again, and the process repeated. At this point, we may be happy with the ranking but we could repeat as many times as necessary (within the time available).
- In four separate, large meetings, we go through the list grade by grade, giving teachers the opportunity to comment on the ranking of students as we go down the list. (This last point added reluctantly – conversations about this stage are ongoing!)