Overview

This cluster-randomized field experiment explores the impact of psychological ownership language in a mail campaign aimed at increasing COVID-19 vaccine uptake among 27,298 residents of a German municipality. Personalized letters were sent to residents, with the treatment group receiving a version that used possessive pronouns to emphasize individual ownership, referring to the vaccine as ‘your’ vaccine, and ‘your’ appointment. Residents’ intentions to vaccinate were assessed through unique click rates to the municipality’s vaccine appointment webpage. The findings reveal that psychological ownership language boosted click rates by 2.5 percentage points or 39%. This study builds on previous research on psychological ownership and further demonstrates the value of this easy-to-implement strategy for influencing willingness to contribute to a public good.

Approach

This cluster-randomized field experiment was conducted in cooperation with the German municipality Bad Nauheim. The experiment involved a mail campaign that sent personalized letters to 27,306 registered residents, grouped within 6,442 addresses. Two versions of the letter were developed, with all letters containing information on COVID-19 vaccines and personalized links and QR codes that directed residents to the municipality’s website for appointment scheduling. Letters sent to the treatment group had subtle language changes in headings and paragraph text that emphasized individual ownership of the vaccine, referring to the vaccine as ‘your’ vaccine, and appointments as ‘your’ appointments.

Addresses were treated as cluster units, with all individuals at the same address receiving the same version of the letter. To ensure proportional distribution of the intervention based on cluster sizes, block randomized allocation was used. Residents were randomly assigned to receive one of two versions of the letter over the weekend of May 21-23, 2021. Intervention effectiveness was measured using unique website click rates for vaccination appointments. The study used linear probability models to estimate the treatment effects. Treatment effects were measured at 1-week and 4-weeks post intervention,finding that use of psychological ownership language increased individual click rates by 2.5 percentage points, a 39% increase, 1-week after the intervention. This effect remained substantial after 4-weeks.

Image Caption: Treatment effects for individual click behavior (1 week & 4 weeks).

Note: ***p < 0.0001.