Bibliometrisk Impact - om konsekvenser ved implementeringen af den bibliometriske forskningsindikator

Abstract

This thesis is about the resent developments in Danish policy concerning higher education: the use of bibliometrics in order to govern funds regarding scientific research. The political idea is to optimize the production of research by rewarding quantity and quality at the same time. The thesis examines the possible outcome of such New Public Management tools, in particular the consequences for the practice of science and research strategies at universities. In doing so, it begins with the 2003 University reform, which created fundamental changes to the organizational structures of the universities, and therefore, presets the conditions in which the bibliometric incentives will work. Secondly, the theoretical foundations for bibliometrics and citation analysis will be explored and traced back to its founding fathers: Robert Merton, Derek Price and Eugene Garfield. These foundations will be questioned through the more resent Actor-Network Theory (primarily Bruno Latour), and it will be argued, that citations can not be said to measure quality, since references are used in more varied ways than the Mertonian assumption suggests. The subject will therefore be analyzed on the basis of interviews with heads of departments at Roskilde University through the application of ANT. The main conclusions are: Publication records, on a departmental/institutional level, will be of greater importance in the recruitment of new researchers. Equally so will the keeping of a continual publication record be of greater concern for the individual researcher. Research in to new research fields could result in periods with fewer publications which could have an effect on department finances. A consequence of this could be the creation of research centers with growing inertia and path dependency, since their risk willingness, most likely, would decrease, as heads of departments and researchers do not want to find themselves in situations where publication gaps could appear.