During the onset of the COVID-19 pandemic, the COVIDiSTRESS Consortium launched an open-access global survey to understand and improve individuals’ experiences related to the crisis. A year later, we extended this line of research by launching a new survey to address the dynamic landscape of the pandemic. This survey was released with the goal of addressing diversity, equity, and inclusion by working with over 150 researchers across the globe who collected data in 48 languages and dialects across 137 countries. The resulting cleaned dataset described here includes 15,740 of over 20,000 responses. The dataset allows cross-cultural study of psychological wellbeing and behaviours a year into the pandemic. It includes measures of stress, resilience, vaccine attitudes, trust in government and scientists, compliance, and information acquisition and misperceptions regarding COVID-19. Open-access raw and cleaned datasets with computed scores are available. Just as our initial COVIDiSTRESS dataset has facilitated government policy decisions regarding health crises, this dataset can be used by researchers and policy makers to inform research, decisions, and policy.

Objective: Although compliance scales have been used to assess compliance with health guidelines to reduce the spread of COVID-19, no scale known to us has shown content validity regarding global guidelines and reliability across a large international sample. Here, we have assessed the validity and reliability of the Compliance Scale developed by the COVIDiSTRESS II Global Consortium, a group of over 150 researchers from across the globe. Methods: We used exploratory factor analysis to determine the most reliable items on the English version of the survey. We conducted a measurement invariance test to determine whether the different language versions of the scale are measuring the same construct with the same measurement structure. Invariance testing indicated that measurement alignment was needed to ensure that the scales are comparable across languages and cultures. After alignment, we employed a novel R code to run MC simulation for alignment validation. Results: We found that alignment of the 6-item Compliance Scale worked well with this method. Confirmatory factor analysis confirmed reliability of the six-item scale. Convergent validity was also found; COVID-19 compliance correlated with vaccine willingness. Conclusions: The Compliance Scale can be employed quickly across multiple languages and countries, and our alignment validation method can be conducted freely in R and employed for future cross-language surveys.

Although compliance scales have been used to assess compliance with health guidelines to reduce the spread of COVID-19, no scale known to us has shown content validity regarding global guidelines and reliability across an international sample. We assessed the validity and reliability of a Compliance Scale developed by a group of over 150 international researchers. Exploratory factor analysis determined reliable items on the English version. Confirmatory factor analysis confirmed the reliability of the six-item scale and convergent validity was found. After invariance testing and alignment, we employed a novel R code to run a Monte Carlo simulation for alignment validation. This scale can be employed to measure compliance across multiple languages, and our alignment validation method can be conducted with future cross-language surveys.