Gold standard guarantees mediocrity

Britain should abandon its costly and cumbersome system of separate centralised national assessments of research and teaching, says David Smith.

BRITAIN has a more costly and time-consuming system of higher education quality assessments than any other country in the world. It also threatens the healthy development of our institutions. The impact of the research assessment exercise on university funding has helped drive up the volume of research to a point well beyond that which existing resources can sustain, hence the current crisis in research funding. Disturbing symptoms such as the growth of "salami slicing" in scientific publications raise doubts about the effect of the RAE on quality of output. Long-term strategic planning in many universities is dominated by the desire to perform well in the RAE. Activities such as providing services to the wider community and writing good textbooks are being put at risk because they do not count in assessments.

The most important aspect of a university for students - the outcome of being taught - defies any meaningful national assessment. About half of all students now graduate with first or upper second class honours, but 20 years ago only a third did so. Nobody believes this shows that quality has improved. Government policy assumes that standards of degree performance should be the same for all universities, yet anyone with extensive experience of external examining knows this is not realistically achievable. Teaching quality assessment has brought some changes, but there is no hard evidence of beneficial effect on student outcome. Preparing for TQA is twice as costly in staff time as preparing for RAE, yet the dominating influence of the latter is such that even the Dearing report concludes that it is devaluing teaching. The correlation between good performance in TQA and RAE is inevitable given the massive effect of the latter on resources (see table), and the way in which research staff and equipment are, quite rightly, used to assist in teaching undergraduates.

All European countries that have introduced systems of quality assessment have firmly rejected the British model. Most follow the procedures of Holland or France, which have institutional self-evaluation at their core; they are simpler, cheaper, less time-consuming and less damaging to staff morale. Funding decisions are not directly and rigidly linked to quality assessments for fear that this will lead to the type of compliance culture that now pervades British universities, and with a consequential negative effect on the delicate mechanisms of real quality management. Continental systems do not measure quality simply by giving numerical values, nor cause the damage that results from two separate and disconnected exercises, each with narrow and restricted interpretations of teaching and research. Instead they enable a coherent integrated assessment of the total activity of a university, including aspects such as international links, local relationships, decision-making structures, planning strategies, etc. My experience of taking part in the scheme organised by the Committee of European Rectors convinces me that the United Kingdom should abandon its present cumbersome and costly system of separate centralised national assessments of research and teaching.

But no abandonment is likely in the near future because assessments are perceived as essential for efficient centralised control of higher education policy. Yet the more this control increases, the less convinced one becomes that it is actually giving society what it needs, and the more one begins to believe that, as in other areas of national life, some exposure of universities to market forces would enable them to respond more easily to real consumer needs.

A good example is diversity of types of institution. All (including government) are convinced of its desirability, yet our system of centralised assessments, especially when coupled with the illusory requirement for a common "gold" standard of performance, is forcing a type of uniformity that will guarantee mediocrity. Government unwillingness to encourage diversity by allowing differing levels of tuition fees seems based on beliefs that it would increase social inequalities. The example of the United States shows this may be mistaken, for its much greater diversity of institutions embraces a more balanced social mix, and the variation in standards and costs is known and accepted. In the widely admired Californian higher education system the state makes no provision for quality audit and assessment and no attempt to impose common degree standards across its various campuses.

Many of the important problems confronting higher education across the world are not solved by assessment systems. All countries face the difficulty of increasing the attractiveness for academic research of teaching relative to research. Many have tried giving prizes for excellence in teaching, but nowhere does this really work. The key requirement is that, just as in research, good performance should be rewarded by academic promotion and its increase in pensionable salary and status. The difficulty of identifying good performance in teaching will become easier not only with the advent of the Institute of Teaching and Learning, but also as machine-based learning spreads rapidly, with teaching packages exposed to scrutiny nationally and internationally.

There is no rigorously proven causal link between the economic prosperity of a country and the numbers of university graduates it produces, but there is a clear link between being a graduate and improved personal earning capacity. This will lead to a general tendency in many countries for the share of the costs of higher education borne by the learner to rise, with corresponding pressure for centralised government control to be replaced by market forces, allowing particular universities to exploit particular niches more effectively. As with other aspects of social welfare, these changes are likely to be sooner in the UK than in most other parts of Europe. This might well prove the avenue for British universities to improve more rapidly than elsewhere. There is an untapped reservoir of entrepreneurial and innovative ideas to be found among British vice-chancellors and their councils and senates, and these could come to the surface to benefit us all if centralised government control was relaxed.

Going to university is an important experience in itself for students, separate from academic learning: it is a time of growing up for school-leavers, a period of freedom and self-realisation for mature students, a time of making unique social contacts and experimentation. When our own children are involved, it matters particularly deeply to us. So thank heavens it is not subject to centralised national assessment.

David Smith is president of Wolfson College, Oxford. This essay is based on the concluding lecture of this year's Wolfson Lecture series (under the title of "The Idea of a University"), to be published by Jessica Kingsley Publishers.