Misfits? Research classification in research evaluation: visualizing journal content within fields of research codes

Abstract

The Australian research evaluation model uses a classification scheme to assign Fields of Research (FoRs) to individual researchers and journals, and to define assessment panels. Eligible journals for assessment are listed and assigned between one and three FoR codes. A high proportion of journals in the list of over 22,000 titles are assigned a single FoR code only. This paper explores the implications of classifying research outputs using the FoR code mechanisms. Eight datasets of title and abstract data from journals assigned a single FoR were mapped using VOSviewer. Four of the datasets were in science fields and the other four were humanities and social sciences fields. The maps and extracted terms for each journal set were examined for overlap with other FoRs. Sizeable overlaps with other FoR codes were observed in the content of three of the sciences fields’ datasets. Weaker overlaps with other FoR codes were found in the humanities and social sciences datasets. The findings suggest that the assignment of FoR codes to journals has the potential to disadvantage researchers and their organisational units.