We reported the number of candidate successes by level and by vocational area as a percentage of those who entered external tests (51 per cent of the total).

We also reported the numbers of students who achieved at least one unit (82 per cent of the total). Commentators were sceptical.

Last year we reported pass rates against numbers of students who had registered. Then why, they asked, did we change the basis of reporting this year?

But no other national exam pass rates are reported against the numbers who begin a course. Registrations for GCSE and A-level are made two or three months before candidates sit exams. Those with little chance of success are not always entered, nor do the figures reflect the individuals who switch programmes after finding themselves on the wrong course. This happens often in FE where many GNVQ students switch programmes early on.

When we reported results against registrations last year they were widely misinterpreted. Some commentators included in the calculations Advanced-level GNVQ students who had only completed the first of a two-year programme.

The presentation we adopted this year helps commentators make a much more valid comparison with performance in academic exams and provides a sound basis for future year-on-year comparisons.

Why, commentators asked, did we report partial successes? Surely those who do not finish a course in time have failed?

Success in GNVQ does not depend on a fixed-date terminal exam. While most full-time students will complete Advanced GNVQ in two years - and an Intermediate or Foundation course in one year - GNVQs are designed to be flexible.

Candidates may choose to take a selection of units only, study part-time or complete the full qualification over a longer period, for many reasons. If candidates enter for GNVQ knowing there is no fixed date when they have to complete, you cannot brand them failures for taking longer than one or two years.

We must also remember that GNVQs are whole qualifications - at Advanced level they are equivalent to two A-levels. This has been confirmed in OFSTED and FEFC inspection reports. If a candidate succeeds in six out of the required 12 units it is reasonable to equate this with an A-level pass. Certainly a student who takes two A-levels but only passes one is reported as a success. Should we not try to do the same for GNVQ?

Finally, we are criticised for the apparent disappointingly low take-up of engineering and technology.

In some subjects, like engineering, construction and IT, GNVQs have only been fully available for a year. So, very few Advanced students have yet completed their courses in these areas. Nevertheless, the number of successes in technology subjects more than doubled this year. While we cannot claim that with GNVQ we have bucked the national trend away from “hard” subjects, we take some comfort from the figures.

None of us involved in the development and delivery of GNVQ is complacent about what has been achieved to date. There is plenty more that needs to be done to ensure that GNVQs are attractive and credible to students, parents, employers and HE. But the figures do speak for themselves - GNVQs are a success story.

Nicholas Carey is chairman of the Joint Council of National Vocational Awarding Bodies