The 2008 RAE should recognise more applied research, but Phil Ternouth remains sceptical.
When the members of the review panels for the 2008 research assessment exercise were announced earlier this month, those of us keen to ensure that the next RAE properly rewards applied and practice-based research scanned them eagerly.
Sure enough, the sub-panel lists do include influential research users.
Microsoft, AstraZeneca, BAE Systems and Rolls-Royce were all there. And although small in number, there is no doubt that there is powerful representation in the more application-related disciplines such as computer science, pharmaceuticals and engineering.
The overall low numbers, however, may be indicative of a continuing bias in the system.
When submitting evidence to the Commons Science and Technology Committee last year, Lord May, president of the Royal Society, famously described the exercise as a "unidimensional totem pole". His point was that it used limited criteria and downgraded the value of applied and practice-based research compared with basic research. As a result, university-business relations, the practical education of graduates, regional economies and the UK's national competitiveness all suffered.
The Lambert review of university-business links also noted that the RAE's focus on basic research limited the desires of researchers to work with business. Analysis by the Council for Industry and Higher Education has shown that business perception of excellence does not always correlate with RAE scores.
In theory, the Higher Education Funding Council for England's attempts to improve the ability of the 2008 RAE to assess "all forms of research on a fair and equal basis" should affect the range of research carried out. But will they?
The retention of the principle of expert review (which the CIHE supports) places considerable responsibility on the experts who lead the panels. In many cases, those experts will have achieved prominence by virtue of their own success in a system biased towards basic research and publication rather than applied and practice-based research and interaction with business.
The system is only now beginning to consider how the excellence and impact of these broader definitions of research might be assessed. It must therefore be open to question whether the panels will respond fully to the Hefce guidance on equal weighting, or whether they will, despite the best of intentions, fall back on the familiar.
Panels may also respond only to submissions made on work done, not work in progress. The deadline for those for RAE 2008 is November 30, 2007. The likelihood is that work is already under way on the research to be submitted. Given the impact of a drop in RAE grading on departmental viability, what is the balance of probability that researchers will be encouraged to "play it safe" and focus on the better understood values of basic research and publication?
The danger is that these issues will instil inertia. If Hefce is serious about improving the RAE, it needs to think now about the extra support it might give to panel members to overcome this.
Given the low numbers of research users on panels, the ability of the panels to co-opt where they feel unable to evaluate niche or interdisciplinary research is very valuable: Hefce must encourage members to do this.
Without firm action from Hefce to demonstrate that the RAE world we knew has indeed changed, the RAE will remain a unidimensional totem pole.
Philip Ternouth is associate director, research and development, at the Council for Industry in Higher Education.