Document is current

Article History

Received: 21 January 2024

Accepted: 13 February 2025

First Online: 19 March 2025

Competing interests

: This research project is part of the National Research Program ‘Big Data’ (NRP 75) of the Swiss National Science Foundation (SNSF) with grant number 166999. Further information on the National Research Program can be found at or . This work has been co-funded by the GFF-IPF Grant of the Basic Research Fund of the University of St.Gallen. We gratefully acknowledge the data access provided by Parship.de (PE Digital GmbH, Hamburg). The NDA between Parship.de (PE Digital GmbH, Hamburg) and the University of St.Gallen grants Parship.de (PE Digital GmbH, Hamburg) the right to access the working paper prior to its release. Parship.de (PE Digital GmbH, Hamburg) was provided access to the working paper prior to its release. Gabriel Okasa is currently employed by the Swiss National Science Foundation, while the presented work has been conducted, when affiliated with the Swiss Institute for Empirical Economic Research, University of St. Gallen. Daniel Boller is currently employed by the World Bank, while the presented work has been conducted, when affiliated with the Swiss Institute for Empirical Economic Research, University of St. Gallen. The views expressed in this paper do not necessarily reflect those of the authors' affiliated institutions.

: The research process was not reviewed by the IRB (University of St. Gallen “Ethical Commission”); an IRB approval is not envisaged as part of the research process. This research does not involve the collection or analysis of data that could be used to identify participants (including email addresses or other contact details). All information is anonymized, and the submission does not include images that may identify the person.

: The data were collected and anonymized by Parship.de (PE Digital GmbH, Hamburg). The platform users agree, as part of the platform registration, that anonymized data can be used for research purposes.