Observing the evolution of the climate change debate across different public spheres
- Monika Rakoczy (Sorbonne Université, Paris, France)
- Robin Lamarche-Perrin and Armin Pournaki (MPI MiS)
Abstract
Climate change is an undeniable truth in the scientific community. However, even though predictions of its impacts have been persisting for years, the debate seems to have gained significant traction in the public sphere only in the last few years.One explanation for this could be the increase of immediate effects of climate change, such as rising temperatures, floods and bushfires.Meanwhile, the increasing interest in the climate change debate could as well have come along with a discursive shift, enhanced by emerging actors on the international scene such as Greta Thunberg. This shift can take place and manifest itself differently in various areas of debate, ranging from political discussions in parliaments to newspaper articles and social media outlets.Therefore, we present our contribution to a climate change observatory currently under development within the ODYCCEUS project. Our focus is to allow the users, in particular social science researchers and data journalists to explore data connected to climate change.
Our contribution consists of accessible methods and Web services to visualize and analyze the climate change debate on different media spheres.
We consider three large-scale datasets spanning from July 2016 to September 2019:
(1) For the perspective of social media, a collection of 80M tweets related to climate change (courtesy of DMI),
(2) For the perspective of mass media, all articles published by The Guardian (courtesy of UVB)
(3) For the political debate, a corpus of 92 287 UK parliamentary speeches (courtesy of MPI MIS)
All documents in these three corpora have been selected by the use of keyword "climate change".
For each of them, we identify and observe the dynamics of subtopics over time, employing methods ranging from outlier exploration and topic modeling to network representations.
This allows us to investigate the possible existence and dynamics of discursive shifts.
In particular, we focus on three aspects of debates: How much is climate change discussed in a given area? What vocabulary / terminology is used to do so? What kind of interaction patterns do we observe in such discussions?
