A global and multilingual portrait of expressed negativity against COVID-19 in social media space

Abstract

The COVID-19 pandemic has increased the incidence of mental health problems in many countries, triggering the public’s unhappiness and sentimental negativity. To date, however, we know little about how sentimental negativity varies across geographic contexts, social groups, and phases of the COVID-19 pandemic. We construct a global and multilingual investigation of the public’s negativity against COVID-19, based on an individual-level sentiment metric via the contents of 2.1 billion geotagged tweets posted from 1 February 2020 to 31 March 2021. Using this tweet dataset covering 217 countries and 67 languages, we further study its dynamics relative to policy implementations, controlled by country-specific mental health indicators employed by United Nations’ World Happiness Report. We find that countries with proper government response and efficient implementation of economic support, and containment and health policy tend to have lower levels of negativity expressed on social media, though the effect of such policies on negativity becomes marginal at the later stage of the pandemic. This study offers a framework for spatiotemporal evaluation of mental health signals to guide through mental health initiatives in response to future pandemics and public emergencies. 

Data Demonstration
  1. Monthly Sentiment Scores by country
  2. Daily Sentiment Scores by country
Data and Code
  1. Source Code