Year: 2022
Type: article
Abstract: There are size-leveling indexes used to demonstrate profiling of entities in different fields, such as the Activity Index (AI) or the Index of Relative Specialization (RESP). Concentrating on the RESP, we consider German state universities as entities and their academic disciplines as fields. While it is common to illustrate several RESP values concurrently using heatmaps, we show that an interpretation of such heatmaps without further information can be misleading. Therefore, we introduce a weight for each RESP value that represents the fraction of a field at a university. Those weights correspond to RESP values that are uniquely identifiable by field and university, resulting in tuples of RESP values and their weights. We introduce a new kind of heatmap that not only illustrates RESP values but represents their corresponding weights. Those new heatmaps are less misleading than classical ones. Our introduction of a new class of heatmaps improves heatmap representation, especially for longitudinal RESP data without the need for additional tables to show the extra information.
Source: Scientometrics
Authors: Joel Emanuel Fuchs, Thomas Heinze
Institution: University of Wuppertal
Language: English
Subfield: Statistics, Probability and Uncertainty
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
Estimated APC paid: $2,780
Funder: Bergische Universität Wuppertal
OpenAlex ID: W4225648484