Arbeitsbereich
Migration und Mobilität
Auf einen Blick
Projekte
Publikationen
Team
Projekt
Studying International High-Skilled Migration Using Large-Scale Digital Trace Data
Daniela Perrotta, Elizabeth Marie Jacobs, Tom Theile, Aliakbar Akbaritabar, Carolina Coimbra Vieira, Emilio Zagheni, André Grow-Böser; in Zusammenarbeit mit Sarah Charlotte Johnson (MPIDR), Ingmar Weber (Saarland University, Saarbrücken, Deutschland), Helga de Valk (Netherlands Interdisciplinary Demographic Institute, The Hague / University of Groningen, Niederlande)
Ausführliche Beschreibung
The study of high-skilled migration is increasingly important as innovation-driven economies depend on the global mobility of talent. Highly educated migrants contribute disproportionately to entrepreneurship and productivity growth, influencing both the competitiveness of destination economies and the human capital dynamics of origin countries. The development of demographic perspectives and theories on high-skilled migration has been hindered by the lack of appropriate data. Digital traces from sources like LinkedIn have opened up new opportunities to measure and explain the gendered dimension of migration aspirations and career opportunities for migrants, the conditions under which skilled emigration leads to “brain drain” or generates “brain circulation,” as well as the determinants and consequences of return migration.
In this project, we retrieve digital trace data from the social networking service LinkedIn to capture the migration dynamics of professionals across Europe and globally. The aim is to leverage LinkedIn data to study high-skilled migration and to complement traditional data sources. Our interests include examining professional migration to new destination countries as well as return migration, a research area that is often understudied due to data limitations. Furthermore, we aim to develop mobility models to assess potential imbalances in migration flows across sociodemographic characteristics and occupational fields.
Research within this project integrates both aggregate- and individual-level digital trace data from LinkedIn. Aggregate-level data come from two sources within LinkedIn: the Recruiter platform and the Advertising platform. While data from the Advertising platform enable us to capture realized migration flows of professionals, data from the Recruiter platform offer us the opportunity to examine intentions to migrate to prospective destination countries. We have demonstrated both the utility and the limitations of relying on Recruiter data as a measure of migration aspirations in Europe (Perrotta et al. 2022). Using gravity-type models, we have identified patterns of relative attractiveness across European countries, highlighting destinations that attract more professionals than anticipated by the models. Moreover, we find significant gender differences, with men expressing significantly greater openness to international relocation than women (Jacobs et al. 2024). In contrast, analyses of realized migrations flows show that women outnumber men among high-skilled migrants in major destination countries such as the United States, Australia, the United Kingdom, and France, particularly in sectors like finance, health care, and real estate.
In parallel, research based on individual-level LinkedIn data enables fine-grained analyses of career trajectories, return migration, and the sequencing of international moves. The data, which contain the CVs of 611 million LinkedIn users, were obtained from a data provider. The data contain information about the users' education and detailed job tenures, including information on the start and end dates of individual jobs, as well as the company and place of each job. We measure migration by observing transitions from one country to another across age, gender, and occupational fields in Europe and globally. To address the systematic biases inherent in LinkedIn data, we have constructed population weights using official statistics, including sources such as Eurostat and IPUMS. This approach enhances the representativeness of our estimates and supports robust inferences on the sociodemographic and occupational differentials in high-skilled mobility worldwide.
LinkedIn data provide insights with high levels of spatiotemporal resolution that would be difficult to obtain through traditional data collection schemes. However, the use of these data comes with challenges, as digital data are generally affected by systematic biases, such as self-selection and measurement errors. While LinkedIn users constitute a sizable fraction of high-skilled workers, they are not representative of the broader population of high-skilled workers. As we refine our approaches to evaluate the causes and consequences of high-skilled migration, we also plan to develop methods to identify the type of information that can be reliably extracted from LinkedIn and other platforms in order to provide generalizations that extend beyond the specific platform.
Offenheit für berufsbedingte Umzüge zwischen europäischen Regionen
Chord diagram of the numbers of LinkedIn users open to relocation between two countries, aggregated to European regions. Direction is indicated by arrowhead and size by the width at the base of the arrow. Proportions have been scaled within each region. © Perrotta, D., S.C. Johnson, T. Theile, A. Grow-Böser, H. De Valk and E. Zagheni (2022). Openness to migrate internationally for a job: evidence from LinkedIn data in Europe. In: Proceedings of the Sixteenth International AAAI Conference on Web and Social Media (ICWSM-22): Atlanta, Georgia and online, June 6th - 9th, 2022; hybrid conference, 759–769. Palo Alto, CA: AAAI Press. (2022)
Migration, Wirtschaft, Erwerbstätigkeit, Ruhestand
Publikationen
Jacobs, E. M.:
Social Forces 103:4, 1538–1559. (2025)
Jacobs, E. M.:
Journal of Ethnic and Migration Studies 51:5, 1346–1370. (2025)
Jacobs, E. M.; Theile, T.; Perrotta, D.; Zhao, X.; Anastasiadou, A.; Zagheni, E.:
Population and Development Review 51:3, 963–994. (2025)
Jacobs, E. M.; Theile, T.; Perrotta, D.; Zhao, X.; Anastasiadou, A.; Zagheni, E.:
MPIDR Working Paper WP-2024-037. (2024)
Sîrbu, A.; Goglia, D.; Kim, J.; Magos, Maximilian P.; Pollacci, L.; Spyratos, S.; Rossetti, G.; Iacus, S. M.:
Journal of Computational Social Science 7:2, 1451–1482. (2024)
Coimbra Vieira, C.; Fatehkia, M.; Garimella, K.; Weber, I. G.; Zagheni, E.:
In: Data science for migration and mobility, 141–158. Oxford: Oxford University Press. (2022)
Perrotta, D.; Johnson, S. C.; Theile, T.; Grow, A.; de Valk, H. A. G.; Zagheni, E.:
In: Proceedings of the 16th International AAAI Conference on Web and Social Media (ICWSM-22): Atlanta, Georgia and online, June 6th - 9th, 2022; hybrid conference, 759–769. Palo Alto, CA: AAAI Press. (2022)
Perrotta, D.; Johnson, S. C.; Theile, T.; Grow, A.; de Valk, H. A. G.; Zagheni, E.:
MPIDR Working Paper WP-2022-007. (2022)