You’ve got mail: increasing vaccination rates

How to cite

Schumacher, Stefanie; Bieberstein, Frauke von; Kulle, Anna-Corinna. 2021. "You’ve got mail: increasing vaccination rates." AEA RCT Registry. AEARCTR-0008070. https://doi.org/10.1257/rct.8070-1.2

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Stefanie Schumacher* , Frauke von Bieberstein , Anna-Corinna Kulle

* Primary investigator

Registry ID
AEARCTR-0008070
DOI
https://doi.org/10.1257/rct.8070-1.2
First registered
Last updated
Status
Completed
Trial dates
2021-06-21 – 2022-09-30
Countries
Switzerland ()
Registry record
http://www.socialscienceregistry.org/trials/8070

Abstract

In light of the COVID-19 pandemic, vaccination campaigns try to further increase vaccination rates. In written reminders, we include behavioral interventions and test their influence on vaccination rates. The study comprises three distinct projects. Project 1 examines the effect of written reminders on vaccination decisions by comparing vaccination rates in our sample to an untreated comparison group. Project 2 tests the effect of social norm information included in the written reminder on the vaccination decision. Project 3 looks at the influence of a mobile vaccination team, offering on site vaccination where no appointment is needed.

Experimental design

We conduct a field experiment, sending out written reminders to get vaccinated against COVID-19 to the inhabitants of several municipalities.

Firstly, the effect of the reminder is analyzed by comparing vaccination rates to a baseline comparison group without reminder. Secondly, the impact of social norm information in the letters is compared between the treatments. Thirdly, in half of the treated municipalities, a mobile vaccination team will be on site and the influence of this offer on vaccination rates is examined.

Design details

Randomization method
Randomization on the household level. The treatments will be randomly assigned using a spreadsheet. In three municipalities, the information on household compositions was not available. In those cases, treatments were randomly assigned on the address-level (streetname + housenumber + postal code). The information on the baseline vaccination rates was not known at the time of the randomization.
Randomization unit
Household level.
Clusters
Across the two experimental and the one control condition in the treated sample, approximately 23’500 households / unique addresses (when household information was not available), equally distributed between all three conditions. For the final sample size, vaccination rates are to be checked.
Observations
In total approximately 46’000 inhabitants receive a letter. Out of this, we expect roughly 40% to be non-vaccinated at the time of our study. This leads us to a final sample of an estimated 18’000 participants.
Arms
We expect approximately 6'000 participants per treatment (in all three treatment conditions). Additionally, we expect approximately the same sample size in our basic (untreated) comparison group.
Primary outcomes
Vaccination rates

Cited by

No citation data available for this registration yet.

Keywords