Evolution of international sentiment towards climate change on Twitter

Abstract

AbstractThe objective of the study was to examine a vast dataset of over 11 million English-language tweets concerning climate change gathered over an eleven-year period. The overarching aim was to illuminate the trajectory and geographical variations in sentiment throughout this timeframe. The authors conducted an exhaustive study of tweets adorned with hashtags related to climate change, using the Valence Aware Dictionary and Sentiment Reasoner (VADER) as a tool to assign sentiment scores and determine the polarity of tweets. Concurrently, it was traced the geographical dispersion of these tweets and their evolution over the duration of the study. The findings unveiled noticeable shifts in sentiment aligning with major global events. For instance, the United Nations' endorsement of the Sustainable Development Goals in 2015 stimulated a rise in positive sentiment toward climate discourse, while the emergence of the COVID-19 pandemic from 2019 to 2021 triggered a significant fall in sentiment scores. The study also detected an ascending trend of positive discourse in the United States and Europe, with Central Africa sustaining the highest average annual sentiment score. On the other hand, the sentiment in New Zealand and India was extraordinarily volatile, exhibiting dramatic changes from one year to the next. In contrast, Australia and New Zealand consistently registered the lowest sentiment averages. Overall, our findings highlight a complex mosaic of sentiment pertaining to climate change discourse across diverse global regions.