Committed Participation or Flashes of Action? Mobilizing Public Attention to Climate on Twitter, 2011–2015

ABSTRACT

Climate advocacy organizations are increasingly focused on citizen mobilization via social media as a strategy, proposing that by mobilizing the existing climate issue public into expressive action they will better be able to pressure policy makers. This paper explores the origins of and rates of participation in bursts of collective attention to the climate issue on Twitter over a five-year period. We find low levels of repeated user participation over time in spikes of public attention to climate on Twitter, while also identifying a small group of organizations and individuals who are “serial” participants in climate actions on Twitter. The high rate of participant turnover suggests continued questions about how to sustain public attention to the climate issue over time.

KEYWORDS:

• Twitter
• activism
• climate
• advocacy
• participation

Disclosure Statement

No potential conflict of interest was reported by the authors.

ORCID

Kjerstin Thorson http://orcid.org/0000-0002-5762-4688

Notes

1 Our goal was to capture the broadest possible set of climate-related tweets, while minimizing the amount of noise in our data. Keyword searches are commonly used to search Twitter datasets (e.g., Freelon et al., 2018). Previous research showed that #climate and #climatechange are among the most common used in tweets about the climate issue (Kirilenko & Stepchenkova, 2014), but also that many tweets about the climate issue do not include the full phrase “climate change” (Thorson et al., 2016), and thus a more narrow search term would result in the loss of a large amount of data. The risk of our approach is that we might have false positives in our data, i.e., tweets that mention the word climate but are not about the climate issue. We analyzed a random sample of data from each event and found very low rate of false positives, an average of 4.25 percent across events (analysis available from the authors).

2 We purchased tweets through Sifter, an authorized reseller of Twitter data.

3 All the usernames are decapitalized for coding purposes. The Twitter search query will also ignore the capitalization issue when searching for usernames.

4 Using the social media monitoring service Crimson Hexagon, we compared the number of tweets containing keyword “climate,” “global warming” and “climate change.” For the keyword “global warming”, the number of tweets increased 2.75 times from 2011 to 2015 (446,115–620,111–694,549–1,078,648–1,260,578 year by year). For the keyword “climate change,” the number of tweets increased 7.26 times in that time period (460,830–765,298–1,055,834–1,875,427–3,345,829 year by year). For the keyword “climate,” the number of tweets increased 6 times in that time period (3,007,191–4,386,663–6,501,427–11,952,996–20,044,344 year by year). In conclusion, we found that the phrase “global warming” is consistently used less often on Twitter than “climate” and “climate change.”

Additional information

Funding

This work was supported by Annenberg Big Data Fund, University of Southern California: [Grant Number None].