ECPR

Install the app

Install this application on your home screen for quick and easy access when you’re on the go.

Just tap Share then “Add to Home Screen”

ECPR

Install the app

Install this application on your home screen for quick and easy access when you’re on the go.

Just tap Share then “Add to Home Screen”

Selective Exposure or Biased Assimilation? Political Polarization and the Sharing of Climate News on Social Media Networks

Immigration
Social Media
Climate Change
Antti Gronow
University of Helsinki
Antti Gronow
University of Helsinki
Arttu Malkamäki
University of Helsinki

Abstract

In the US, the rise of partisan media is connected with increasing political polarization where political attitudes move further apart on the political spectrum. The main mechanism connecting the consumption of partisan media with polarization is selective exposure, which refers to choosing political messages that reinforce existing attitudes. In a partisan media context, the reception of climate-related news also tends to become a partisan issue. However, we know a lot less about countries where media-party parallelism is not such a big deal. A case in point is Finland where news consumption is not as divided along partisan lines as in the US. We argue that in this context biased assimilation may play a bigger role than selective exposure in explaining how news media relates to political polarization. Biased assimilation refers to a setting in which the same news stories are seen by actors of different political leanings but their interpretation differs based on political ideology. We study the role that sharing news media links plays in the polarization of climate discussions on Finnish social media. First, data were collected on X with keyword searches that cover the years 2015-2023. Next, the users were divided into three ideological bubbles based on how they reposted content produced by politicians of different ideological leanings. Then, bipartite networks in which the users are connected if they post the same news media links, were projected onto unipartite networks and analyzed using the adaptive E-I index to quantify selective exposure as the ratio of links being shared within and between bubbles. Third, we evaluated the sentiments of the users posting news media links with the help of a multi-billion parameter language model. If different ideological bubbles share the same news links but with different sentiments (negative vs. positive), this counts as evidence of biased assimilation. We also compared the sharing patterns of links to climate news with the topics of immigration and COVID-19. We find that although both selective exposure and biased assimilation have increased, especially during 2019-2023, the most prominent climate news receives considerable attention across the bubbles and is consistently interpreted through their ideological lenses. Thus, the role of biased assimilation should not be downplayed in explaining the association between media consumption and political polarization. In addition, we note that the sharing patterns of links to climate news closely resemble those to immigration news, suggesting alignment of the two topics. Our study contributes to the study of political communication and the way in which news media contributes to the polarization of the discussion on social media in a non-US context.