We’re all in the very difficult position of on the one hand knowing that any grade boundaries we create are a fiction of some degree or another but on the other hand still being expected by our schools, academy chains and, ultimately, students to say where they are in relation to the new 9-1 grades.
My own academy chain has worked with several HoFs to devise grade boundaries, but these differ by as many as three grades from the grade boundaries provided by PIXL (ours are harsher). To continue the fiction metaphor, it seems like it just depends whether we want a fairy story or a horror story depending on which grades we use.
However, AQA has offered a service where schools can upload their marks onto something called Testbase Camp and this has provided population data on the most recent sample assessment materials provided for the English Language exams.
Using these and a few other pieces of information from here and there I’ve put together the following:
Grade Boundary Rationale Using Testbase Camp
This is the process that grade boundaries for AQA English Language were arrived at using the population data from Exampro Testbase Camp published 13th January 2017.
Data on for the national population distribution of each paper and combined can be provided, but it is accessible from Testbase Camp itself.
Guidance from “Setting the grade standards of new GCSEs in England – part 2”
“Calculating 1-7
- Broadly the same proportion of students will achieve a grade 4 and above as currently achieve a grade C and above in the subject.
- Broadly the same proportion of students will achieve a grade 7 and above as currently achieve a grade A and above in the subject.
- Broadly the same proportion of students will achieve a grade 1 and above as currently achieve a grade G and above.
- Grades 2, 3, 5 and 6 will be awarded arithmetically so that the grade boundaries are equally spaced in terms of marks from neighbouring grades. Grade 5 will be positioned in the top third of the marks for a current grade C and the bottom third of the marks for a current grade B. “
“Calculating 8-9
- Across all subjects (as opposed to within each individual subject) close to 20 per cent of those awarded a grade 7 or above will be awarded a grade 9 (the ‘tailored approach’).
- Under the tailored approach, grade 8 will still be awarded arithmetically so that the grade boundary is equally spaced in terms of marks from the grade 7 and 9 boundaries. “
JCQ Percentages 2016
http://www.jcq.org.uk/examination-results/gcses/2016/gcse-full-course-uk-by-age-2016
| Grade | Cumulative Percentage 2016 |
| A | 11.9 |
| B | 33.3 |
| C | 63.4 |
| D | 84.6 |
| E | 93.5 |
| F | 97.5 |
| G | 99.1 |
Using “Setting the grade standards of new GCSEs in England – part 2” to transfer from letters to numbers.
| Grade | Cumulative Percentage | Comments |
| 9 | 2% | 20% of Grade 7 |
| 8 | 7% | ‘Arithmetically equally spaced’ (7 and 9) |
| 7 | 12% | Grade A 2016 to integer |
| 6 | 26% | Grade B 2016 – ((Grade B 2016 – Grade A 2016)/3); 33.3 – ((33.3 – 11.9)/3) |
| 5 | 43% | Grade B 2016 + ((Grade C 2016 – Grade B 2016)/3); 33.3 + (63.4 – 33.3)/3) |
| 4 | 63% | Grade C 2016 |
| 3 | 75% | ‘Arithmetically equally spaced’(1 and 4) |
| 2 | 87% | ‘Arithmetically equally spaced’ (1 and 4) |
| 1 | 99% | Grade G 2016 |
Comparative Raw Mark Grade Boundaries
| (Out of 80) | Academy Chain | PIXL | TestbaseCamp |
| 1 | 11 | 4 | 5 |
| 2 | 18 | 10 | 23 |
| 3 | 28 | 16 | 30 |
| 4 | 41 | 22 | 35 |
| 5 | 49 | 29 | 43 |
| 6 | 56 | 36 | 50 |
| 7 | 64 | 43 | 57 |
| 8 | 69 | 50 | 61 |
| 9 | 76 | 56 | 63 |
Ultimately we will only know in August how correct these are, but for the time being, as far as I’m aware, these are the best we have to go on. It does, of course, assume that the marks provided to Testbase Camp are correct, as they would not be subject to external moderation.
I hope this is useful and if there are any flaws in my reasoning, logic or just maths, please let me know!
Unto the Breach 