Institutions and the map of science: matching university departments and fields of research

It is an increasingly common practice within universities to use departments as units of research funding and there exists in the form of the extensive British Research Assessment Exercise (RAE) a national funding system which is essentially tied to departments as units of analysis. Yet the interdisciplinary nature of modern scientific research, where researchers in departments publish in journals across a range of fields outside their nominal disciplinary affiliation, is an acknowledged 'norm' in the university research community. This paper uses complete data for all Australian universities to explore the correspondence between the designations of departments and the designations of the fields and subfields to which members of these departments contribute through their publications. Previous studies of this aspect of knowledge production have centred primarily on micro-level data relating to particular specialities and departments. We suggest that the use of performance indicators at the level of university departments inevitably obscures important features of modern research. Any attempt to introduce a system-wide evaluation of research based on the university department would have particular disadvantages for interdisciplinary research and for those newer institutions which have not organised their academic structures along traditional departmental lines. We suggest in relation to research funding bodies, either internal or external to the university, that there should be an increased use of field-coded research information. (C) 1998 Elsevier Science B.V.