The COVID-19 pandemic was the first in which social media became a popular platform for expressing public opinion and spreading misinformation. While attempts have been made to understand the public’s adherence to protective behaviours during the pandemic, few studies have utilised social media data. Furthermore, few studies have sought to investigate whether there is any link between their calculated behavioural metrics and COVID-19 outcomes of interest. In this presentation, I will discuss the utilisation of Twitter data to extract proxy measures for adherence to protective behaviours. I will also present preliminary results of sentiment analysis conducted on a collection of policy-specific terms in the UK. Additionally, I will explore whether these results are related to COVID-19 outcomes of interest.
Invited Speaker: Tristan Naidoo (Imperial College London)