Preferences Expressed Through Twitter

Preferences, attitudes, and well-being affect people's economic decisions and their effective adherence to policies. Arguably, the COVID-19 pandemics changed these factors, but we do not know the direction of the changes nor how permanent they are. Today more than ever, decision makers need timely information on these factors to effectively introduce policies to exit from the crisis, promote economic recovery, and support social cohesion. This project addresses this urgent need by informing about how people's preferences, attitudes, and well-being changed during the COVID-19 crisis, and whether such changes are permanent or transitory. We use sentiment analysis on data sourced from Twitter to provide real-time tracking of the social change triggered by the pandemics. In this way we aim to provide timely information and to avoid delays typical of conventional large scale surveys. We will track the changes of life satisfaction, mental stress, trust in others and in institutions, loneliness, uncertainty about the future, populism, and attitudes towards globalization. The project will produce a time-series database for each of these aspects in Luxembourg and 5 European countries severely affected by the crisis (Italy, France, Germany, Spain, and United Kingdom). The data will go from December 2019 to December 2020. We believe that timely data releases produced by our project will be a useful tool to inform economic and social policies, and an important contribution to public and scientific debate at European level.