Migrants vs. stayers in the pandemic – A sentiment analysis of Twitter content

Abstract

In this paper, we propose a sentiment analysis of Twitter data focused on the attitudes and sentiments of Polish migrants and stayers during the pandemic. We collected 9 million tweets and retweets between January and August 2021, and analysed them using MultiEmo, the multilingual, multilevel, multi-domain sentiment analysis corpus. We discovered that the sentiment of tweets differs between migrants and stayers over time, and it relates to the country of migration. The general sentiment is similar for migrants and stayers, but a more detailed analysis reveals that hashtags related to staying safe and staying at home, as well as vaccinations are more polarised for migrants than for stayers, and they reflect the general development trend of the pandemic in Europe. In addition to comparing migrants with stayers, we also compared migrants staying in different countries. amongst the countries of migration, for which we collected at least 3000 tweets, the most positive sentiment of Polish migrants’ tweets was observed in Belgium, with the most negative sentiment coming from Estonia. We also observed that the sentiment of tweets written in Polish by stayers in Poland is less negative when compared to Polish migrants in most of the countries with the highest number of tweets.

Keywords

• Sentiment analysis
• Text mining
• Text analytics
• Social media
• Twitter
• Migrants