Twitter provides a useful tool for studying public conversations about climate change, an issue that crosses international boundaries and stirs political and scientific debate. This review presents what is currently known about the way climate change is discussed on Twitter, acknowledging advantages and limitations and suggesting future areas for study. As an accessible platform, Twitter allows public expression of opinions on climate change and provides data on how these fluctuate over different times and places. Moving forward, studies assessing climate views can be improved by better linking them to demographic and other data indicating the population that Twitter users represent. The open‐ended content of tweets provides additional information, such as which topics are associated with climate change and which terms are used to discuss it. Future studies can build on these results to capture a wider range of climate‐related discussion. Finally, researchers are using Twitter to understand who initiates and participates in climate change dialog and how, by categorizing users as different actors—such as politicians, celebrities, NGOs, or the general public. Future research could consider how effective these efforts are, and how Twitter activity translates to offline outcomes.

The general public’s perceptions of climate change may be shaped by local climate impacts through the mechanism of experiential processing. Although climate change is a long-term global trend, individuals personally experience it as weather from moment to moment. This study assesses how New York State adults’ overall perceptions of their personal experiences with the effects of climate change and extreme weather (surveyed in early 2014) are related to recent weather conditions. This research is unique in that it examines multiple types of weather: temperature and precipitation, over 1 day or 1 week, quantified both as relative and nonrelative measures. Respondents’ perceptions that they had personally experienced climate change or extreme weather significantly increased with warmer relative (percentage of normal) minimum temperatures on the day of the study. Maximum temperatures and total precipitation levels were not significant predictors of perceptions of personal experience, either on the day of the study or over the preceding week. Experiential processing had a smaller effect on perceptions than motivated reasoning, the influence of preexisting ideas. Respondents who believed that climate change was happening agreed more that they had personally experienced it or extreme weather, and this effect increased for individuals who thought that climate change was anthropogenic, as opposed to naturally caused. Of the sociodemographic factors assessed here, political party, gender, and region were significant predictors, while age and education were not.

This study examines tweets ( N = 1,103,119) sent by 638 US politicians regarding climate change ( N = 10,135). Using data reflecting the risks faced and opinions held by their constituents, the extent to which politicians lead—tweeting about climate change for those most at risk, called trusteeship—or follow—tweeting about climate change for those who are already concerned, called playing to the crowed—is examined. Results reveal that while Democrats generally tweet about climate change more often than Republicans, within each party, tweeting frequency is mostly explained by the level of concern in a politicians’ constituency. Objective risks faced by constituents play no role. Analysis of behavior across different office levels—federal, state, and local—also shows that while federal politicians are more partisan, state and local politicians play to the crowd to a greater degree. Analysis of the tweets shows the politicians’ unequal engagement leads to over-representations of some topics (e.g., “calls for action”) and under-representations of other topics (e.g., criticizing fossil fuels and deniers).

No abstract