Like the Department for Education and Employment, Bob Doe seems to believe that performance-related pay (PRP) for heads and deputies will work; that schools will improve more quickly with it than without it (TES, December 6, 1996).

As a statistical consultant in industry I have tried to keep abreast of the argument about PRP - and, as I understand it, the research evidence shows it to be puzzlingly ineffective. And it is clear from anecdotal evidence that its unintended side-effects can be horrendous. For a sceptical view from a “management guru”, see Douglas McGregor’s book The Human Side of Enterprise. For an all-out assault, see almost anything written by W Edwards Deming.

Whether a governing body or a local authority is responsible for setting senior staff pay, some sort of quality test must be applied as the increments use public money. In theory, the existence of any such test means that we already have PRP. But, in practice, there is a world of difference between trying to avoid rewarding the incompetence of a small minority and trying to use pay as an instrument to improve the work of the vast majority. It is the latter which leads to disappointments and side-effects.

Nevertheless, the bandwagon of PRP for headteachers seems unstoppable. So how can we avoid the worst of the pitfalls? One essential condition is to ignore the DFEE “guidance” pointing towards outcome measures like exam or test results and attendance rates, because: league tables mean these measures already get more than enough attention; improved outcomes are usually the result of actions taken over a period of years (possibly by the previous job-holder); there are many reasons why results may vary, up or down, completely outside the control of the person being assessed; and most important of all, only a small fraction of a school’s overall objectives produce measurable outcomes.

In principle, governors should be in a better position than local authority staff to take a wider view. It is their responsibility to ensure that the school has a broad set of aims or objectives. They are also, in principle, much closer than local authority staff to the processes by which those aims are delivered. They could be trained to assess whether a few key processes (which might be identified in the school development plan) have been improved, based on their own observation and feedback from other groups (classroom teachers, non-teaching staff and pupils, where appropriate). This would be the “previously agreed criterion” for the assessment of the head’s performance.

In contrast to Terry Mahoney’s view (TES, September 20) that assessing performance is a job for experienced professional personnel managers, I believe that the kind of procedure I have outlined would be far better than the appraisal process found in most organisations.

But would governors be able to find the time (including time for training)? Or will the pressure of time combined with pressure from the DFEE combine to squeeze out the more thoughtful alternatives? I am not optimistic.

Dan Grove is a co-opted governor in Hampshire.