Erroneous exercises 1

Why do we persist with the expensive, time-consuming and blunt instrument of the research assessment exercise in subjects that have research councils ("Break up RAE scam", THES , September 12)?

When I submit a proposal to the Particle Physics and Astronomy Research Council, it is assessed by a panel of five scientists in my subfield of astronomy, who send it out for review by three experts on the specific topic of the proposal. Their recommendations are scrutinised by an overview panel of four, who weigh the proposal against those from other areas of astrophysics.

In the 2001 RAE, there were just two astronomers on the panel, leaving huge areas of astrophysics with no "front-line" representation. Intense peer review by research councils could be regarded as overkill. But according to the Higher Education Funding Council for England's own research performance indicators, there are large variations between institutions in the ratio of research council support to Hefce support. Even among "traditional" universities, this varies by more than a factor of two.

The different funding outcomes of the research council and Hefce processes suggests that at least one of them is badly flawed. I know which one my money is on.

Michael R. Merrifield
University of Nottingham