Institutional cooperations in Austrian research: An analysis of shared researchers

Abstract Researchers increasingly hold multiple organizational affiliations, yet the effects of such arrangements on organizational performance remain poorly understood. We develop a scalable network-based framework that represents simultaneous affiliations as relational links and apply it to harmonized bibliometric data from Austria. Two coaffiliation networks are constructed: a network capturing all simultaneous affiliations and a temporally filtered network retaining recurring affiliation pairs. Analyzing these networks in parallel allows us to distinguish transient from persistent coaffiliations and examine how short-lived and enduring ties shape organizational connectivity. Network regressions show that geographic proximity remains an important determinant, with greater distance reducing the likelihood of a coaffiliation link forming. Once transient affiliations are removed, most same-type organizational links lose statistical significance, indicating that many observed coaffiliations reflect temporary overlaps rather than durable structural ties. Persistent cross-type links are rare, with ties primarily observed between universities and research institutes. To evaluate the effects of coaffiliations on organizational citation impact, we use field- and year-normalized indicators. Sensitivity analyses show that citation impact depends on how affiliations of multiaffiliated authors are assigned, including whether the first or last listed affiliation is used. Research institutes and universities exhibit higher citation impact, and persistent coaffiliations are associated with greater visibility.