A few years ago, a school that was looking for a new head invited candidates to a “rotating dinner” at which guests had their starter on one table, their main course on another, and dessert on a third. The idea was to give them a chance to talk to everyone involved in the selection decision. It was an awkward affair - and as a way of seeing how candidates behaved in a normal social situation, it failed miserably.

Governors often bring out the sherry and nibbles, or even dinner plates, when appointing a head - because social events can be a good way of introducing candidates to a school. But as a selection tool, “trial by sausage roll” has serious flaws. It raises suspicions that people are judged on their social graces, whether they are “one of us”, and not on their ability to do the job. As the governors who hosted the rotating dinner found, trying to structure such an event fairly can make things worse.

“If you have no structure at all, you can’t really use any evidence from it. But the further you go down the road of putting structure on a thing like that, the more farcical it becomes,” says Maureen Cooper, a director of the education personnel consultancy EPM Ltd.

The selection of headteachers often includes informal interviews with inspectors and advisers, and in-tray exercises to test problem-solving skills and the ability to prioritise. In the majority of cases, candidates are also asked to make presentations, and the centrepiece of the process is usually a formal panel interview. Bringing in up to a dozen people to interview candidates is meant to ensure that decisions are based on a variety of viewpoints. But don’t such interviews just show which candidate is best at performing in front of a panel? The business world, where such an approach is now out of favour, seems to think so. And it seems schools are also losing faith.

Panel interviews certainly won’t be used to choose the head of the new Compton Enterprise and Sports Academy, due to open on the site of an existing school in west London in September 2003.

“It’s an exceptional job and we can’t afford to make mistakes,” says Alec Reed, chairman of the Reed employment agency group, which is sponsoring the city academy to the tune of pound;2 million.

Candidates for the job, which has a salary of pound;100,000 plus perks, must show that they have turned a school around. They will also undergo psychological tests to see whether they have the qualities of leadership, teamworking and innovation. Finally, there will be one-to-one interviews with people who have a track record as successful school leaders.

“At panel interviews there’s a tendency to follow the chairman, whereas I’d like to have everybody’s distinct opinion on each candidate,” explains Mr Reed.

The Secondary Heads Association, which offers a consultancy service on headteacher appointments, hasn’t gone as far as abandoning panel interviews, but the association’s consultants advise governors that rather than one large panel, they should set up several small ones to look at various aspects of the job, such as leadership, professional competence, and communication skills.The association also offers schools an assessment centre service, whereby candidates on the final shortlist are tested against a set of competencies. Widely used outside education, assessment centres give recruiters a chance to see how applicants behave in group discussion and simulations. In a recent survey by the Association of Graduate Recruiters, these came out as the most useful and effective recruitment tool. But they are also expensive.

Of the 50 or so headteacher appointments which the association’s consultants handle each year, only a few involve assessment centres. But according to Harvey Black, the consultancy’s principal officer, three days of interviews, discussions and presentations give interviewers a good handle on what candidates are like. Selecting people for teaching jobs usually takes up less time than choosing heads and deputies. Where schools struggle to find staff, they sometimes appoint overseas teachers after no more than a quick telephone interview. But more often, candidates are given a tour of the school and an interview, although in this case big panels are now less common, too.

Many schools also ask candidates to give a demonstration lesson. This can show their general approach to teaching and how they relate to pupils - but only up to a point. As one of many messages on this subject posted on the TES website puts it: “I deal very much in rapport in the classroom to get the best out of kids. Going in cold was not going to help.”

Richard Taaffe, a former headteacher in Barking and Dagenham, north-east London, and now head of recruitment and retention at the TimePlan teacher supply agency, believes that ‘demo’ lessons identify good showmen and women rather than good teachers. He says that a better way to find out whether candidates are right for the job is for prospective colleagues to take applicants around the school, talk to them and at the end of the day report their impressions to the head.

“If they wholeheartedly want someone, then they are really going to help that person out when they struggle - and everyone struggles in a new job,” says Mr Taaffe.

Another model is for recruiters to visit candidates in their current schools and observe them at work. This is obviously impossible with overseas teachers or those who are not in a job, but for most teaching jobs, and even management posts, this is probably one of the most effective selection tools. And it certainly beats the sausage roll test.