Tracking author affiliation drift: A matrix-based method for identifying temporal patterns
Journal
Journal of Informetrics
Journal Volume
19
Journal Issue
4
ISSN
1751-1577
Date Issued
2025-11
Author(s)
Abstract
This study presents a matrix-based framework for tracking and classifying researcher affiliation drifting, with a particular focus on multi-country co-affiliations. By structuring author-affiliation data into time-sequenced matrices, the method captures both the persistence and configuration of institutional ties within individual publications. Each paper is categorized based on the types of co-affiliated countries, and researchers are subsequently classified into field-independent typologies reflecting the degree and structure of their institutional mobility. Applied to a dataset of Highly Cited Researchers (HCRs) in mathematics, the framework reveals notable affiliation patterns—most prominently, a high concentration of researchers exhibiting simultaneous affiliations across multiple countries without transitional or exploratory affiliation types. These observations demonstrate the method’s utility in surfacing affiliation structures that may not be visible through conventional bibliometric indicators. While the mathematics domain serves only as an implementation example, the results echo broader concerns about the strategic use of multi-affiliations in certain fields. The proposed approach contributes a replicable, scalable tool for analyzing affiliation dynamics, with implications for bibliometric research, institutional evaluation, and science policy.
Subjects
Affiliation drifting
Affiliation matrix
Affiliation typology
Author affiliation
Multi-affiliation
SDGs
Publisher
Elsevier Ltd
Type
journal article