Thousands of students who flunk school GCSEs are condemned to failure by being put on the wrong college courses, Further Education Funding Council inspectors warn.
A report this week catalogued "a depressing picture" of repeated failure as colleges allow too many young people to resit GCSEs, instead of taking vocational studies.
The report of a national achievement survey questioned whether the college curriculum has place for GCSEs, which are designed as a school-leaver certificate.
In 1995, 145,000 school-leavers and adults sat GCSEs in colleges. Only 9 per cent of 16 to 19-year-olds gained grades A*-C in five subjects. Another 9 per cent gained four top grades and more than 20 per cent failed to achieve a C or above in any subject.
Sixth-form colleges got the best results. Adults who were at school before the introduction of GCSE did much better than school-leavers who had recently failed. "For many of them, the repetition of GCSEs reinforces failure and decreases motivation," the inspectors said.
Jim Donaldson, FEFC chief inspector, said: "Colleges need to do much more to stimulate and re-engage these students in learning. They need to look at how they manage their GCSE programmes and they need to look more critically at the pre-enrolment guidance they are giving some of their applicants."
The report also cites poor attendance and completion rates on GCSE resit courses. There are no detailed figures available, but they would give an even bleaker picture if taken into account.
At A-level, figures now being collected by inspectors are said to give a truer picture of student achievement than annual exam league tables.
While league tables compare A-level passes with the number of students entered, the new data also includes students who drop out.
A Portsmouth College report showed that 91 per cent of those entering English language A-level passed. But, measured against enrolments, the pass rate fell to 69 per cent.
Portsmouth is one of the first colleges to have such figures published as a table in its inspection report. Such comparative tables at A-level are expected to become an increasing feature of inspections.
Mark Griffiths, a senior FEFC inspector, said: "Improved quality assurance systems allow colleges to compare exam passes with the number of students who start the course."
Figures which were published last year showed that two-thirds of students who complete courses such as A-levels gain a final qualification. Gathering data to compare success with enrolments has been tougher.
Mr Griffiths said: "If you look at the initial enrolments, you get an accurate picture of retention over time. "
As GNVQ students do not have to be entered for a final exam, the number of awards made each year is often compared with initial registrations two years earlier.
City amp; Guilds director general Nick Carey said: "Unfair comparisons have been made with A-levels. I am delighted we are now getting like-for-like data which puts GNVQ figures into context."