Zum Hauptinhalt

Overcoming COVID-19 Vaccination Resistance When Alternative Policies Affect the Dynamics of Conformism, Social Norms and Crowding-Out. Proceedings of the National Academy of Sciences of the United States of America

Universität Konstanz; Santa Fe Institute, New Mexico

So schwer die Pandemie in Millionen Leben auch eingegriffen hat: In Deutschland gibt es keine Pflicht, sich gegen SARS-CoV-2 impfen zu lassen. In anderen Ländern ist eine solche Verpflichtung längst eingeführt, sei es für Pflegekräfte in Italien oder für Studierende in Kalifornien. Werden sich in Deutschland genug Menschen freiwillig impfen lassen, um die Pandemie zu beenden?

  • Disziplin: Sozial, Wirtschaft
  • Forschungsmethode: Quantitativ
  • Forschungsdesign: Primärerhebung, Repräsentativ angelegte Befragung
  • Erhebungsstatus: Erhebung abgeschlossen, Ergebnisse veröffentlicht, Daten zugänglich

Ziele der Studie

Der Bereitschaft zur freiwilligen Impfung und möglichen Reaktionen auf eine Impfpflicht sind die Verhaltensökonomin und Psychologin Dr. Katrin Schmelz und ihr amerikanischer Kollege Prof. Dr. Samuel Bowles nachgegangen. In der repräsentativen Online-Befragung „Leben im Ausnahmezustand“, die der Exzellenzcluster „The Politics of Inequality“ der Universität Konstanz durchführte, fragten sie nach dem persönlichen Impfwunsch: Würden die Befragten sich freiwillig impfen lassen wollen? Und würden sie sich immer noch gern impfen lassen, wenn eine Impfpflicht ihnen die Entscheidung aus der Hand nähme?


To identify differential individual responses, all subjects were asked to state their agreement to get vaccinated in both cases: if it remains voluntary and if it is enforced. In a separate survey, the scientists investigated the possibility of a demand effect due to asking a subject to answer both questions. They implemented a between-subjects design confronting respondents with only one alternative (either voluntary or enforced) and obtained very similar results. Altering the order of the alternatives in a within-subjects design did not affect average agreement with enforced or voluntary vaccination either.

To limit a potential spillover effect — a subject answering questions in a way to minimize cognitive inconsistency — the module containing the questions on agreement to get vaccinated and the module containing the questions about vaccine effectiveness and perceived restriction of freedom in case of enforcement were separated by a module unrelated to vaccination.

The questions were embedded in an ad hoc online survey on COVID-19 initiated by the Cluster of Excellence “The Politics of Inequality” at the University of Konstanz.

The predefined target sample size was 4,700 subjects in the first wave, and aimed at 60% of those first-wave participants for the second wave. Participants were recruited from a commercial online access panel administered and remunerated by the survey provider respondi, which usually conducts market research. Membership of the respondi survey pool and participation in its surveys is voluntary and follows a double opt-in registration process. Participation is incentivized with tokens that can be exchanged for goods. Given this material incentive, people registered there are unlikely to have atypical intrinsic or social motivation relevant to the subject matter of the survey. This is important because otherwise, voluntary participation in the survey might create a sample bias in favor of voluntary policies.

The panel was implemented and run by the surveyLab at the University of Konstanz. The first wave was conducted from April 29 to May 8 and the second wave from October 28 to November 6, 2020.


The anonymized survey data and code files to replicate the results of the paper have been deposited at GESIS SowiDataNet datorium (German Data Archive for the Social Sciences) and are available at https://doi.org/10.7802/2272