Longitudinal data collection on an international level during the COVID-19 pandemic

Abstract

Background Different sampling approaches can be used in longitudinal surveys. The COVID-19 pandemic disrupted possibilities to conduct random based longitudinal surveys. However, longitudinal data on public mental health responses are urgently needed. Methods We conducted a longitudinal study in eight European countries (Albania, Belgium, Germany, Italy, Lithuania, Romania, Serbia, Slovenia) including 18+ year olds with the goal to reach as many people as possible and generate robust data. We used various sampling approaches (convenience sampling, snowball sampling, panel data) utilizing social media, professional organisations, educational institutions and personal direct contact. To gather robust data, focus was on multiperspectivity rather than representativeness. A mix of sampling approach is possible. Digital sampling methods were feasible in the European region. Results In total 2,426 participants filled out >50% of the questionnaire in the overall sample, which is, also due to the rapidness of survey development, a good result for data on mental health and resilience during COVID-19 pandemic. COPERS study has shown that a mix of various sampling approaches is possible and a good option to generate datasets in a relatively short time span. However, no responses rates can be calculated using. Conclusions Online surveys are a good tool to reach people in times of crisis. Furthermore, they can be adapted directly in case of changes, if necessary and desired. Using different ways of recruitment, a diverse group of people can be recruited. However, online surveys require a certain technical understanding, which not all potential participants have. Question remains how multinational research is evolving and being shaped. This asks for possibly new methods to define and detect sampling bias.