The paper presents the results of two Bradford analyses conducted on two different types of journal articles produced by departments at Uppsala University, Sweden. The two types of journal articles studied are “refereed” and “other (popular science, discussions, etc.)”. The results show that the rank ordered lists of departments
vary a lot, and thus that results of Bradford analyses are depending in part on the types of journal articles included in the study. The results are discussed and connected to problems and challenges related to concept operationalization.