Some companies worry about high staff turnover; others clearly wish they had more. Organisations such as fast food outlets and insurance companies learn to live with fast staff turnover as a matter of course.

But when an organisation begins to feel it is losing its best people, haemorrhaging its talent or simply experiencing a sudden loss from particular departments or sections, questions have to be asked.

Most managers have their own theories as to why staff leave. These may include poor pay, lack of career advancement or simply a poor selection decision in the first place.

But others are not so sure and conduct some informal research. This usually involves little more than an interview with those who voluntarily quit rather than being fired or made redundant.

The theory is that such interviews can simultaneously fulfil a number of functions. They supposedly help an organisation to modify causes of attrition and other festering corporate problems.

They can help some people because they give them a chance to let off their anger or disappointment in front of someone who is willing to listen.

In this sense the exit interview is more about damage control and prevention.

At the most mundane level, exit interviews are about "leaving protocol", including the orderly and sensible recovery of the employer's property (passes, badges, etc) and ensuring that buildings and accounts remain secure.

So how do they work? The exit interviewer is usually chosen from the human resources department rather than from line managers. Ideally, they will be benevolent, avuncular "good listeners" who take the side of the departing employee.

They will be trained to act like clinicians and counsellors, being non-judgmental and focused entirely on the employee and his or her experience.

They have a set of standard, open-ended questions, which include: "How did you feel you were managed during your time with us?"; "In general, how do you feel this organisation is run?"; "What could have persuaded you to stay with us another five years?"; "What did you enjoy most/least about your time with us?"; "How would you rate the work climate/corporate culture?" and, lastly, "What are the main motives for your leaving?"

What is interesting about these questions is that they should all have been asked earlier - probably by the boss during an appraisal.

The fact that no one asked the questions is no doubt a partial indicator of why the person has chosen to leave.

Interviewers should expect the unexpected emotional outburst: rage, tears, shouting. They should take rough notes, including whether they would hire the person if they reapplied for a job; or, more poignantly, what lessons could be learnt for hiring others in the future.

Part of the agenda of the exit interview is to understand the leaver's view of his or her compensation and benefits package and to influence it where this would be appropriate.

The factors influencing departure are well-known: unhappiness at the place of work, combined with the attraction of another job.

Factors cited include quality of supervision; relations with co-workers and subordinates; workload; job security; flexibility of hours; salary and benefits; location/commuting distance and personal or family matters.

From the employer's viewpoint the idea of the exit interview is eminently sensible. In fact, it benefits the employer much more than the employee. It can gather useful information, soothe psychological wounds and prevent people turning into whistle-blowers and "terrorists" who may subvert the organisation's reputation after they have left.

But from the point of view of the job-leaver, it has few benefits. It is tempting, when faced by a sympathetic senior person in the organisation, to unload the frustration and fury one has built up over the years.

The release of emotions may feel cathartic at the time but it is probably ill-advised in the long run. It is unwise to burn your bridges if you are in a small world, town or sector: remember, the organisation takes notes on what you say at the exit interview.

The best piece of advice is twofold: dissimulate or keep quiet.

Do not say that your work was humdrum, stressful or tedious . . . talk of moving on to new challenges to upgrade your portfolio and extend your horizons.

Human resources practitioners are told that a good exit interview may help to limit costly legal action and other bad publicity instigated by a disgruntled employee. Remember: the notes of the exit interviewer could always be Exhibit A in some file somewhere.

One of the main problems of the interview is timing. People in the grieving business talk of a wet and a dry ceremony. The burial/cremation is full of shock and tears; the gathering to celebrate someone's life, held six months later, is more measured and more balanced.

The emotional stress of leaving (even in the most happy circumstances) can lead to distorted, unreliable exchanges from both parties.

Resentment overcomes reason; emotion overcomes evaluation. The grieving process takes time and people can reflect more accurately on what they miss, have lost and feel. Thus it may be best to have an exit interview between six and nine months after the person has left. They may be invited out to, or back to, lunch to discuss the issues the company wants to know about.

The advice for employees remains the same but from the company point of view it means they may gather some really useful intelligence from the delayed exit interview if they know what they are doing.

The author is professor of psychology at University College, London