Michael Gove believes in evidence-based policy. He is desperate to learn from international surveys of school performance, but now the education secretary's own EBac is providing vital evidence on what has been happening at home.
Heads resent the EBac's sudden imposition and question its composition. But the measure sheds important new light on the murky world of secondary school exams.
Academies have been among the EBac's most vociferous critics, and it is easy to see why. A TES analysis last month showed they were significantly more likely than other secondaries with similarly disadvantaged intakes to fail on the measure.
Collectively, academies have been reluctant to reveal exactly which exams they use. When think-tank Civitas surveyed 118 academy heads in 2009, only 55 per cent believed breakdowns of their results should be published, and only 43 per cent released them.
Since then anonymous responses, parliamentary questions and EBac figures have combined to slowly lift the veil.
We now know that in 2009 only 52 per cent of academies' GCSE results comprised actual GCSEs, compared to 76 per cent for other state secondaries. And academy pupils were significantly less likely to be entered for GCSEs in geography, history or modern foreign languages.
Some argue that the equivalent qualifications that academies tend to use instead - condemned by critics as "pseudo-vocational" - are useful and motivate pupils. Mr Gove is not among them. He told Parliament he was "worried about the use of so-called equivalent qualifications instead of GCSEs" and he has excluded them from the EBac.
Yet he continues to back a huge expansion of the academies, particularly academy chains, which make the most use of these equivalent qualifications.
Evidence suggests they are not "among the very best performers in GCSEs", as Mr Gove has claimed, but often use alternative routes to league table success.
A new division in secondary education has quietly emerged, with many pupils from deprived backgrounds steered towards different qualifications to their middle-class counterparts.
It would be unfair to say only academies are doing this. But the EBac figures suggest they are more likely to than other schools faced with the same challenges.
The chains claim it takes time to achieve EBac success. But many of these academies have been open for at least four years and still have miserable scores. One says it inherited teachers who needed training to teach GCSEs. But academies enjoy unprecedented freedom over staffing.
Whether the EBac is a fair measure of success remains a moot point. But if Mr Gove believes in it he must surely take notice of the evidence it presents.
ACADEMY CHAINS - How they score
EBac scores (percentage of pupils achieving measure) in six major academy chains:
- Harris Federation - GCSEs sat in eight academies last year. EBac scores of 0, 1, 2, 3, 4, 6, 11 and 32 per cent.
- Ark - GCSEs sat in five academies. EBac scores of 0, 0, 2, 4 and 5 per cent.
- Haberdashers' - GCSEs sat in three academies. EBac scores of 2, 6 and 32 per cent.
- ULT - GCSEs sat in 19 academies. No pupils at all achieved EBac in six of them. Scores of 1, 2, 2, 4, 6, 6, 7, 7, 9, 11, 16, 17 and 42 per cent in the other 13 academies.
- E-ACT - GCSEs sat in eight academies. EBac scores of 0, 1, 1, 1, 2, 2, 3 and 16 per cent.
- Oasis - GCSEs sat in 10 academies. No pupil achieved the EBac in four. Scores of 1, 3, 3, 5, 8 and 10 per cent in the other six.