Just because you know what a car is doesn't necessarily mean that you're competent to drive! It's with good reason that you have to pass your driving test before you set off solo behind the wheel. And yet, in universities "the doctrine remains wholly ingrained that gaining a degree is sufficient for teaching at degree level and that holding a PhD is all but sufficient for doctoral and research supervision," wrote Andrew Morgan in a letter to the Times Higher Education Supplement published on 7 September.

Morgan is right to point out the absurdity of this situation, but perhaps the doctrine is becoming less ingrained than he thinks. It was announced last month that the Biotechnology and Biological Sciences Research Council (BBSRC) is to extend its Training and Accreditation Programme for Postgraduate Supervisors (TAPPS), which has run as a pilot scheme for the last 3 years at the Institute of Animal Health, to universities and institutes throughout the UK. And less than a week after the publication of Morgan's letter, a session at the Society for General Microbiology's (SGM) meeting at the University of East Anglia addressed the serious issue, 'Research Supervision: How to Get it Right'.

Speaking at the SGM meeting, Howard Green, dean of the graduate school at Staffordshire University and chair of the UK Council for Graduate Education pointed out that "supervision and research are separate skills", and that a good researcher is not automatically a good supervisor. He said that there are several reasons why universities are now waking up to the concept of supervisor training and development. These include external factors, such as the Quality Assurance Agency's Code of Practice for postgraduate research programmes, which states that "Supervisors should have the necessary skills and experience to monitor, support and direct research students' work," and adds that "Institutions should consider the provision of training for supervisors and continuing staff development."

Another factor pushing institutions to take the quality of supervision more seriously is the increasing tendency of students to complain if they feel they're getting a raw deal, and even to sue. But James Groves, general secretary of the National Postgraduate Committee (NPC), believes that those postgraduates who actually give voice to their concerns represent only a small minority of the aggrieved students. Indeed, a survey of 1000 postgrads carried out in 1997 found that 61% of those questioned had experienced problems with supervision, but only 18% had pursued a grievance. There are many reasons why students may chose not to say anything. They may not want to rock the boat because they are relying on their supervisor for a reference. Or they may be discouraged from saying anything because their supervisor also holds a departmental position, such as postgraduate tutor, head of department, or dean of research, which means that they would also play a key role in the complaints procedure. It's crucial, said Groves, that students are "able to discuss problems in confidence and informally," and if their supervisor is also in a position of authority, they must be provided with an alternative route to air their concerns. Providing such a safety valve should allow problems to be diffused before they reach the courts.

The all-important supervisor-student relationship is the most difficult aspect of the PhD for both parties, it would seem. Adrian Eley, a microbiology lecturer at the University of Sheffield, believes it is very important that any supervisor development programmes that are introduced be aimed at the needs of both supervisors and students. As part of his dissertation for an MEd in teaching and learning, he surveyed both groups and found that this was one area in which they were in total agreement! Asked to list the most important areas for supervisor training, both supervisors and students placed the supervisor-student relationship as the top priority, followed by 'managing research'; they put 'monitoring student progress' in third place.

So if both students and supervisors can see areas where training would help, it seems a pity that so few opportunities are available. Green knows of only five formally taught programmes of research supervisor training. One is at Bradford University, which has adopted an institution-wide approach and insists that all new members of academic staff take its course. It's a key element in Bradford's plan to upgrade the quality of its research, explained Green, and significant effort has gone into overcoming the natural cynicism of the academic staff--including making all drinks at the bar free for the first half hour of each evening session!

The TAPPS scheme, which has now been extended to all institutions that want to adopt it, was begun in BBSRC-funded research institutes, because they are not attached to higher education institutions and so it was necessary "to be sure that researchers develop these skills," explained Green. The TAPPS course is delivered through a series of workshops covering such essentials as 'effective recruitment and selection of students' and 'coaching and learning styles', but online support and mentoring is seen as important, too. Indeed, alongside TAPPS, the Gateway on Research Supervision has been developed to point supervisors to useful resources that will help them in their quest to become better advisors.

TAPPS is entirely voluntary, but supervisors who take part do work toward accreditation by submitting a portfolio of written evidence to one internal and one external assessor trained and appointed by the TAPPS committee. Nevertheless, the twin issues of whether such training should be compulsory and whether competence should be assessed at the end of it remain vexed. It may be easy enough to insist that all new staff members undergo some training, but "what do you do with the mid-50s, grey-suited professors?" asked Green. And while accreditation is quite a popular concept ("Increasingly people want something at the end to say they've done it," says Green) it could lead to embarrassment. There's no point in assessment unless failure means something. But "if an elderly professor fails, have any of us the courage to say that he can no longer supervise?" Green wondered.

Meanwhile, Eley's survey has turned up one problem for both supervisors and students, a problem that could be fixed relatively easily. 'Finding time for supervision' was one of the most difficult problems cited by supervisors. But with an average of almost five students per supervisor among those questioned, it is hardly surprising that supervisors find this a difficult juggling act, particularly when undergraduate teaching and administrative chores also have to be factored into the equation. Perhaps learning to say 'no' to the temptation to squeeze just one more student into the lab could result in a better all around experience for everyone.