Citations and Their Use as Indicators in Science Policy. Studies of Validity and Applicability Issues with a Particular Focus on Highly Cited Papers Dissertation

This paper presents the results of a study in which scientists were asked about their own publicationsand their citation counts. The study shows that overall the citation counts of the publicationscorrespond reasonably well with the authors’ own assessments of scientific contribution. Generally,citations proved to have the highest accuracy in identifying either major or minor contributions.Nevertheless, according to these judgements citations are not a reliable indicator of scientificcontribution at the level of the individual article. In the construction of relative citation indicators theaverage citation rate of the subfield appears to be slightly more appropriate as a reference standardthan the journal citation rate. The study confirms that review papers tend to be much more frequentlycited than other publication types. Compared to the significance authors attach to these papers theyappear to be considerably “over-cited”. On the other hand, only marginal differences are found in thecitation rates between empirical, methods and theoretical contributions.