Welcome to the OSF-Page of our article. ### Important links Find the full published article here: https://rdcu.be/cJIZN The accompanying website is found here: https://digitalemuendigkeit.github.io/covid-19-behavior-sem All code is found here: https://github.com/digitalemuendigkeit/covid-19-behavior-sem Anonymized data are found here: https://osf.io/bkwau/ (CSV and RDS format). ### Abstract In the ongoing COVID-19 pandemic, non-pharmaceutical protective measures taken by individuals remain pivotal. This study aims to explore what motivates individuals to engage in such measures. Based on existing empirical findings as well as prominent behavioural the-ories, a partial least squares structural equation model (PLS-SEM) of predictors for pandemic protective behaviour was estimated using a representative German sample (n=437). The study was preregistered at OSF. The model explains 69% of the variance for behavioural intention, which is strongly correlated with behaviour (ρ=0.84). The most influential pre-dictor for protective behaviour is its perceived efficacy, followed by normative beliefs and perceptions about costs for protective behaviour. Distrusting beliefs in science and scientists negatively predicted response perceptions and were also strongly and negatively correlated with behaviour. Knowledge about COVID-19 was weakly linked with perceived response efficacy, as well as with behaviour. These findings suggest that communication strategies surrounding COVID-19 should emphasise the efficacy of responses and foster a sense of responsibility. ### Raw Data availability Reasonable requests to access the raw data can be directed to either Lilian Kojan or André Calero Valdez.