Double RAE ratings

The improved consultation and transparency that accompany the forthcoming research assessment exercise are more than welcome.

The consultation highlights a major problem, however. It seems that the primary determinant of an RAE grade will be the quality of published work submitted - whether it is of international, national or sub-national quality.

But panels will also assess things that in many departments will be only weakly related to the quality of their published work, such as the number of research students, financial value of research contracts and grants awarded and whether a departed colleague has been replaced.

We urgently need to know, panel by panel, if such ancillary criteria can actually affect a department's grading -that is in the worst case, obscure recognition of top-quality work. But surely the most sensible way to solve the problem of measuring two incommensurable things would be to give each department two ratings: one for the quality of its publications and one for the general context in which research is conducted. In some cases, this could have the highly desirable effect of spotlighting those departments that are good but not very well supported by their institutions.

Andrew Fleming Department of archaeology University of Wales, Lampeter