Risk perception during the COVID-19 pandemic

In this project, our research group and colleagues study risk perception and optimism about the current COVID-19 pandemic. The project was started in mid-March, just before the lockdowns in Germany, the UK and the USA. A first preprint is available and a second preprint, in which we will discuss our longitudinal data collected in the following months, will be published within the next weeks.

Using online platforms (soscisurvey and prolific), we collected people’s responses about various measures related to risk perception to address the following questions: 1) how do people perceive the risk of getting COVID-19/infecting others/getting severe symptoms/etc. 2) how do those perceptions change over time; 3) what factors influence risk perceptions, and 4) do those risk perceptions predict adherence to protective measures. We used a longitudinal design, re-inviting the same participants 3 times (mid-March, early April, mid-May). Each time, we also collected independent datasets to replicate and confirm our findings. Data were collected in the UK, USA, and Germany, three countries with different trajectories of cases and governmental strategies, thus allowing cross-country comparisons. We will continue collecting data to study the long-term changes in these measures.

We find that people show comparative optimism with respect to getting infected with COVID-19 relative to someone like them. This optimism bias is stable over the first 2 months, despite the absolute perception of risk reducing over that time: in the beginning (in mid-March) most people thought there was a medium to high chance that they would get infected or infect someone else with COVID-19, but 2 months later most people showed a reduction in the perceived risk of getting infected or in infecting someone else. Surprisingly, the optimism bias (comparison between oneself and someone similar), remained stable during this time.

Comparing countries revealed few differences in risk perception but large differences in how much people trusted their government to handle the situation well. While in German there was strong trust, there was mixed trust in the UK and low trust in the US. Crucially, these differences between countries in trust in their respective governments seem specific to the government and aren’t present for trust in science.

We are currently finalising the analysis to reveal to what extent risk perception predicts adherence to protective measures (physical distancing, and hygiene measures such as hand washing and mask-wearing).

Find the preprint here: https://psyarxiv.com/epcyb/

This project is run by Benjamin Kuper-Smith, Christoph Korn, and Lisa Doppelhofer from the DNHI lab in collaboration with Gabriela Rosenblau (George Washington University) and Yulia Oganian (University of California San Francisco).