Mapping climate discourse to climate opinion: An approach for augmenting surveys with social media to enhance understandings of climate opinion in the United States

similarity within a topic and contrast between topics [26]. There are several ways to compute coherence and one of the most widely used metrics is Cv [26]. This metric combines several measurements, specifically, indirect cosine measure, boolean sliding window, and normalized pointwise mutual information, and has been shown to accurately indicate the degree to which a topic can be easily interpreted [26]. To determine the ideal number of topics, we iteratively construct models over a large range of topics with a step of two and computed coherence with the Cv metric. Based on these results, we select the range of topic numbers where coherence was highest and repeat the process with a smaller range and step size of 1. Many topic models constructed with Twitter data have over one hundred topics in the final model [20]. However, as the tweets used in this analysis already have a prioi filtering related to climate change, we use lower ranges of topics as the training corpus is unlikely to contain as many topics as an unfiltered tweet stream. In the first round of model development, we test models with 10, 12, 14, . . ., 50 topics. Based on these results (depicted in S2 Fig in S1 File), we then test models with 14, 15, 16, . . ., 22 topics. Based on the coherence score analysis, we identify seventeen as the appropriate number of topics in the model. This set largely encompasses the breadth of Twitter discussion on climate change while minimizing overlap between topics. The final topic model makes use of the entire

350,000 tweet corpus to ensure that the training data accurately represents the extent and vari- ety of discourse on Twitter. A concise summary of the final topics is presented in Table 1. It should be noted that the topic names and descriptions (Table 1) are subjective and repre- sent our best effort at concisely describing the complexity of each topic category. A key chal- lenge in LDA analysis is discerning the distinct characteristics of each topic set. The state-of- the art involves referring to the top n-grams and most representative tweets for each category (see S3 Table and S4 Fig in S1 File)—an inherently subjective task. While this analysis is based on quantitative representations of these topics, it is critical to acknowledge the subjectivity inherent in unsupervised topic models and its implications for the study’s inferences. Table 1. Categories from LDA topic analysis and a brief summary of their content. Topic Description Climate Impacts Adverse impacts, such as sea level rise, drought, forest fires, etc. Earth Day Celebration, activities, and tips related to Earth Day Politicians Political platforms of democratic candidates and the US President Sustainability Promotion Advice on reducing carbon footprint and greenhouse gas emissions Activism Climate activists (e.g. Greta Thunberg) and events (e.g. Extinction Rebellion) Green New Deal Commentaries on Congress’s Green New Deal proposal Them, Not Us How others are affected by and should respond to climate change Clean Energy Progress towards renewable energy targets and technologies

Environmental Justice Concerns over environmental degradation, especially the impact of pipelines Youth & Education Role of parents and teachers in educating young people about climate change Climate Denial Arguments supporting the denial of climate change Grassroots Action Social media based efforts to solicit signatures for various petitions Carbon Tax Discussion of relationship between taxes and climate change Weather Reports Daily weather updates issued via Twitter Global Warming Arguments supporting global warming as a natural phenomenon Trust in Science Discussion of the importance of science in decision making Plants, Trees Relationship between plant life and climate change https://doi.org/10.1371/journal.pone.0245319.t001 PLOS ONE An approach for augmenting surveys with social media to understand climate opinion in the United States PLOS ONE | https://doi.org/10.1371/journal.pone.0245319 January 14, 2021 6 / 16