PETER KNIGHT has produced an informative and thought-provoking analysis of comparative funding of teaching between universities in his article "The funding stakes" (THES, November 15).

While there is much in it of value to those involved with funding, the article should carry the following health warning. These figures are potentially dangerous. They should not be used: * by administrators, as a substitute for "expenditure per FTE student".

* by government, as a measure for ranking the teaching efficiency of different universities.

* by students, as an indication of the level of resourcing of courses at different universities.

³ÉÈËVRÊÓƵ

Without such a warning there is a real risk that the figures on which Dr Knight bases his analysis (Average Units of Council Funding) will be taken as an analysis of the total funding for teaching at each university per full-time equivalent student. They are not. They include only that funding which comes from the ³ÉÈËVRÊÓƵ Funding Council for England's teaching grant and exclude funding from all other sources, eg student fees for home, European Union and especially full-fee overseas students plus contributions from other funds which a university may decide to channel into teaching.

At the University of Surrey, for example, the funding council's teaching grant accounts for only 40 per cent of income attributable to teaching. This percentage is likely to vary widely between universities, depending in part on deliberate decisions taken within a university to attract students who bring different levels of fee income but who do not attract extra teaching grant.

³ÉÈËVRÊÓƵ

This fact limits the valid application of figures, such as Dr Knight's, which measure only one part of the funding mix, the teaching grant.

A university which has chosen to diversify its funding will be able to devote a higher level of expenditure to teaching each of its FTE students than the figure shown as its AUCF.

The AUCF for such a university, if misused, would give the impression that it is either more efficient than it really is, and/or less generous in its provision of resources.

Other factors which restrict the valid applications of AUCF are the simplifying assumptions that the student body mix at all universities (eg ratio of types of full-time students) is identical and so too is the mix of disciplines within academic subject categories. In the absence of other appropriate data, there is a real danger that analyses based on AUCF will be used for purposes other than those for which they are valid. Anyone, seeking to make decisions which require, explicitly or implicitly, comparisons of expenditure per FTE student, efficiency of teaching or provision of resourcing who reaches for the bottle labelled "AUCF" should be aware that it is unlikely to be effective for their symptoms and will probably exacerbate their problem.

³ÉÈËVRÊÓƵ

By highlighting the risks inherent in the mistaken application of AUCF, we hope to encourage the work required to produce that information to make valid comparisons. It is an important aid to those universities striving to improve further and it deserves measurements which are fit for the purpose.

Bob Gunn

Director of planning

University of Surrey