The Dynamics of Digital Environmentalism: Computational response to environmental policy, events and stimuli

Lead Research Organisation: University of Oxford
Department Name: Sociology


This research presents a methodologically innovative means of analysing online environmentalism and understanding information flows. Using advanced quantitative methods, this research combines multiple methods to address the aforementioned questions; applying large-scale Twitter data in a novel manner through comparison with time-series trend data, as well as application and connection to both media and survey data. This provides an opportunity for cutting-edge research, and as a result, this
research allows for potential collaboration with environmental NGOs and activist groups, or organisations such as The Behavioural Insights Team, Twitter and Google.
The research will primarily involve analysis of web data and network coding using time-series methodology, which will highlight direction and precedence of topics and information flows. To better understand the dynamics of bottom-up activism on social media I will conduct discourse and hashtag analysis surrounding different environmental events and topics. As Twitter has open access, I will use Barberá et al.'s (2015) approach and look at hashtags and keywords associated with the environment and associated policy, to see which has the highest engagement over time. This is made possible through accessing Twitter's public API. To do so, I will use the ESRC COSMOS Open Data Analytics software, which facilitates the breakdown of Twitter data by time, topic and geolocation, whilst allowing for both frequency and network analysis. I will also assess the diffusion of the posts by analysing who the content was shared by, shown by the number of retweets (González-Bailón, Borge-Holthoefer & Moreno 2013). Within this, I will additionally look at top-down social media usage, for example government and media accounts. Overall, this will allow me to see how certain events are To compare online activism with the dynamics of mainstream media reporting on environmental issues I will analyse historical media data. Through Carbon Brief and other organisations who systematically gather news articles on environmental topics, or alternatively via the LexisNexis news archive, I will measure variance in online reporting; using text coding to measure type and number of newspaper articles, and how such levels shift over time. To measure broader online environmental interest and news consumption I will use Google Trends data, which provides information on Google searches for particular terms. The trends data is able to show comparison of interest for particular terms by location and time period, as well as how some search terms rise and fall in popularity in conjunction with others. This will allow for a better understanding of when environmental issues gain traction with the broader public more generally. In doing so, I will compare this against dates of real-world events, to see how environmental searching for particular topics varies with other exogenous occurrences such as policy implementation, climate changes and natural disasters.


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
ES/S50158X/1 01/10/2018 31/03/2022
2267096 Studentship ES/S50158X/1 01/10/2019 30/09/2022 Martha Radford-Kirby