The percentage of students expected to graduate from different English universities varies from 48 per cent to 98 per cent, according to a report by the National Audit Office.
Improving Student Achievement in English Higher Education , published today, found high overall levels of achievement but wide geographical variations.
"Overall, achievement rates are impressive," the report concludes. "Some 77 per cent of full-time first-degree students were projected to achieve a degree at the institution at which they started, 1 per cent to achieve a different qualification and 5 per cent to transfer to another institution."
The report says that this compares well with other countries. Only the Japanese have better achievement rates - but they have a lower participation level. The report looks at completion and achievement - from selecting courses to achieving qualifications and employment prospects.
But it warns that universities may be in danger of missing the government's learning targets. "To achieve government targets, institutions will need to encourage wider participation, maintain standards and raise achievement rates," the report says. Higher education, it says, should provide different and more resource-intensive support for students struggling to achieve their qualifications.
Jeff Jones, director of education value-for-money studies at the NAO, said:
"More than 30,000 students starting full-time first-degree courses each year fail to get a qualification. This can involve significant personal cost, both emotional and financial."
The NAO examined the projected achievement rates published as part of the Higher Education Funding Council for England's performance indicators last month. The University of North London predicted that 48 per cent of its full-time students starting first-degree courses in 1998-99 would graduate. Cambridge projected a 98 per cent success rate.
Hefce produces a benchmark for each university and assesses performance indicators against it. Benchmarks take into account differences between universities - such as students' prior achievement, the balance of young and mature and subject mix. "Although variations appear to be wide, performances tend to be close to benchmark," the NAO report says.
The NAO also found great variation in continuation rates, defined as those students who continue beyond their first year. "Some institutions lose only 1 or 2 per cent of their students during the first year, while others lose more than one in five," the report says. "Among mature students, the variation is even wider, with a small group of institutions losing up to a quarter."
Students who withdraw tend to have lower qualifications and are more likely to have come through clearing. There was very little correlation with social class. A-level results were by far the clearest indicator of academic success.
Students on medical sciences, education, languages and humanities courses tended to have better continuation rates than those studying engineering, technology and mathematical or computer sciences.
The NAO interviewed 26 students on why they dropped out. They identified five main reasons: a lack of preparedness for higher education; changing personal circumstances or interests; financial matters; the impact of undertaking paid work; and dissatisfaction with the course or institution.
Mr Jones said: "Many academics argued that the demands of the national curriculum left pupils with too little time for structuring their own learning, and consequently they found university a struggle."
The NAO also found wide variations in graduate salaries. Male graduates earn about 17 per cent more than male non-graduates and female graduates 30 per cent more than female non-graduates. Graduates from social class IV (partly skilled) and social class V (unskilled) earn on average 7 per cent less than graduates from social class I (professional) or social class II (managerial and technical) backgrounds, the report says.
"A 7 per cent return on earnings, plus the debt associated with higher education, must make certain social groups question the financial benefit of higher education," Mr Jones said.
As well as analysing data for the report, the NAO set up focus groups, interviewed staff, students and former students, surveyed all institutions on their management practices and visited six institutions.
Focus groups highlighted the need for better initial descriptions of courses. "We were shocked to learn that some students had only learnt late in the day that their course was not accredited by the professional body of the industry that they sought to join," Mr Jones said.
The NAO also found that students were reluctant to apply for discretionary funding, complaining that the application procedure was complicated with no guarantee of success.
The NAO recommends:
* Comprehensive information about courses
* Additional guidance and information for students who come through clearing
* Publicly available data on graduate destinations and employment beyond six months after graduation
* Better mechanisms for rewarding most effective lecturers
* More action to identify those students who may benefit from extra academic support
* Students should have a regular schedule of meetings with their personal tutors.