Ideological biases in social sharing of online information about climate change

of the sceptic domains. Taken together, these findings demonstrate significant strength of polarisation and politicisation in information-sharing about climate change on Twitter, with two large clusters of users and information sources, broadly characterised as a left-wing/envi- ronmentalist group and a right-wing/sceptic group. Characterisation of information-sharing communities The main source domains and indicative content of media articles shared within the five larg- est communities in Week 4 can be seen in Fig 6, in which the radii of the circles show the rela- tive sizes of these major communities. Here Communities 1, 3, 4 and 5 are communities within the left-wing/environmentalist cluster in the information-sharing network, whereas Community 2 is the single community in the right-wing/sceptic cluster. Fig 6 uses a TF-IDF weighting scheme for each token, such that the prominent tokens are those that are character- istic of a particular community when compared to the network as a whole. Source domains are shown in Fig 6a and content is shown in Fig 6b (unigrams) and Fig 6c (bigrams), to allow a characterisation of the broad themes in each community in terms of geographic focus, political or climate science biases and key subjects of interest. Table 2 summarises these communities using the data from Fig 6. Looking at source domains (Fig 6a), different geographic and political biases can be inferred based on earlier analysis of domain ideology. Community 2 features predominantly right-wing sources, whereas Communities 1 & 3-5 contain content from left-wing sources.

Fig 4. Climate media content is politicised. Mean political ideology (left-to-right) and climate opinion (environmentalist-to-sceptic) expressed in content from the 62 coded web domains over the six coders (see Ideological coding of source domains). Point size is proportional to the square root of total share count and lines indicate ± one standard deviation. Labels are shown for 10 most frequently shared domains in the coded list. https://doi.org/10.1371/journal.pone.0250656.g004 PLOS ONE Ideological biases in social sharing of online information about climate change PLOS ONE | https://doi.org/10.1371/journal.pone.0250656 April 23, 2021 13 / 25