ML numbers at AL and GCSE, and “severe grading”

I've prepared a summary of the latest situation regarding the numbers taking ML at A-level and GCSE (+ spreadsheet with all the numbers and graphs, updated to include June 10 results), and also regarding “severe grading”.

I’ve updated the pages on the ALL London site so that these documents can easily be downloaded. I have created 5 separate pages with up-to-date figures:

- numbers at GCSE ML over time
- numbers at AL and AS ML over time
- severe grading at GCSE
- severe grading at AL and AS
- the issue with the new A* at AL for ML

There is also a timeline relating to the last few years, with presentations, documents etc, including the presentation I gave at the Ofqual Inter-subject comparability seminar in Oct. ‘08

The issue with the new A* in A-level French (and to lesser extent in German and Spanish) is a new one, and worth pursuing vigorously before it becomes embedded. It is also one which is unlikely to have been noticed at an individual school level, and only becomes clear looking at the national statistics.

There are also links to:


BBC News article today about Ofqual’s brief to look at comparability of qualifications (but not the points equivalence).

BBC News article “GCSE results: Trends explained” - “CILT, the national centre for language learning, agreed with this [John Dunford’s] analysis, saying the trend was "less to do with student disaffection" and more to do with "performance table pressures".

“Rule of thumb”

Judging from the ML forums, the numbers and results in ML continue to be source of much anxiety and concern, incl. how they are judged. A further thought to help fair judgement would be to publicise a “rule of thumb” arising from the QCA report (Feb ’08): if all pupils took Fr (or Gn) and Maths, to get the cohort % who “should” get A*-C:

Take the Maths A*-C cohort rate e.g. 70%
take the Maths C grade percentage e.g. 30% ( and so 40% had A*-B), and halve it e.g. 15%
Add this to Maths A*-B rate e.g. 40% + 15% = 55% to get the French cohort %

Clearly if ML is optional, then this needs to be adjusted, esp as those choosing ML are likely to be better at it, but at least it is a quick “rule of thumb”

I hope the above help to give a factual basis for the situation.