Quality is about more than cash 1

In the discussions on the Treasury's bombshell proposal to replace the research assessment exercise with a metrics-driven procedure, a number of important issues have been overlooked (News, April 7, 14).

One is the dual-support system for research. In this, quality-related money sustains the basic infrastructure as well as staff - while the research councils and other bodies support specific projects.

According to the proposals, this grant and contract income from research bodies will then determine the QR allocation. Isn't this inverting the dual-support model? You get infrastructure money only when you have attracted grant money, which assumes you already have the infrastructure.

The proposed system will inevitably focus QR money on those institutions that already have a "superior" infrastructure: if you haven't got the foundation for attracting research grants, you will never be able to invest in it.

Second, the proposal is a blueprint for concentrating power within institutions. The metrics are to be applied to universities as a whole, with no separate assessments of individual departments. With the RAE - for all its faults - we are told not only the perceived quality of departments but also how much QR they earn for their universities, which gives them a bargaining chip in the intra-institution contest for resources. Without such information, managements will be free to rely on their own biases in deploying money: the fate of chemistry at Exeter and Sussex universities may be a harbinger of more closures.

The documentation for RAE 2008 clearly states that the criteria for judging research quality must be "originality, significance and rigour". Research income is a poor indicator of those criteria - the most original and significant research does not necessarily require the largest sums of money. Judgments about research originality, significance and rigour are subjective, but they are best made by academic peers. The proposal to replace such judgments by research income metrics trivialises the RAE. Peer review will be replaced by the political whim of secretaries of state and vice-chancellors, and the future of knowledge production placed in their hands.

Ron Johnston
Bristol University