The majority of students think academics are good at explaining things, make the subject interesting and are enthusiastic about what they teach.
But they are less impressed with the feedback they get, and many find the workload overwhelming and unrealistic.
The first findings from a pilot of the National Student Survey, involving almost 15,000 students and graduates from 23 higher education institutions in England, Wales and Northern Ireland, reveal that students are more positive than negative about teaching quality in each of the seven areas tackled.
Ministers believe the survey is a significant step towards improving student choice, offering comparative information across institutions and helping to reduce dropout rates.
Alan Johnson, the higher education minister, said that one of the biggest reasons for students dropping out was that they had chosen the wrong course.
"A national student survey will give students better information to help them make informed choices about where and what to study at university. It will also enable them to look at the quality of the course alongside the tuition fee to be charged for it," he said.
But Mandy Telford, president of the National Union of Students, warned:
"What the government must not do is use these surveys to create university league tables to justify some institutions' desire to charge much higher top-up fees."
Ms Telford said that the NUS supported the introduction of a student satisfaction survey but, for this to work, there had to be guidelines and a representative sample of students questioned.
The pilot's findings are being analysed, and a second pilot will report by summer. A full-scale national survey is due to be launched in 2005.
Findings show that students rate overall quality highly and give the biggest thumbs-up to teaching and improving generic skills, with an average of 3.9 out of a possible 5. Most students feel they have gained confidence in tackling unfamiliar problems and have improved their communication skills. Assessment and learning resources gained 3.7 out of 5. Learning support gained 3.3, and university departments were awarded 3.2 for feedback and workload.
The pilot compares anonymised institutions in business studies and social studies. The average score across 20 institutions for teaching quality in social studies ranged from 3.7 to 4.2.
Comparisons between five business studies courses found similarities on workload levels and learning resources but significant variations in the quality of feedback and levels of support.
The survey also reveals diversity within institutions. One institution had an average score of 2.5 for the quality of feedback in business studies but scored 3.8 for the course's quality of teaching.
The pilot report says it is still assessing the variation in feedback, since some of this may not stem from the quality of teaching. Instead, it may be due to the individual students involved in the survey and also to the "experimental" nature of the pilot, which used different survey methods at different institutions and achieved "highly variable response rates".
Hot lecturers and nasty ideologues