In “News and Social Media Accurately Measure Protest Variation”, Anton Sobolev, Keith Chen, Jungseock Joo, and Zachary C. Steinert-Threlkeld show that news and geolocated social media data generate accurate estimates of protest size variation. This claim is substantiated using cellphone location data from more than ten million individuals during the 2017 United States Women’s March protests. These cellphone estimates correlate strongly with those provided in news media as well as three size estimates generated using geolocated tweets, one text-based and two based on images. Inferences about protest attendance from these estimates match others’ findings about the Women’s March.

In “How to Use Social Media Data for Political Science Research” Pablo Barbéra and Zachary C. Steinert-Threlkeld offer an overview of existing research that uses social media data in the fields of Political Science and International Relations. It discusses two types of studies: those where social media is being used as a source of data to study, e.g. political networks or public opinion, and those focusing on how social media transforms different political phenomena, ranging from ideological polarization and misinformation to election campaigns. Our review also offers an in-depth analysis of the main challenges of this type of data, such as the different sources of bias that limit the generalizability of findings, the difficulty of connecting online and offline behavior, and concerns about reproducibility and ethics. To illustrate the opportunities and limitations of social media data, we also provide an applied example using Twitter as event data to study the dynamics of protest movements in Egypt and Bahrain in 2011.

In “Social media and Russian territorial irredentism: some facts and a conjecture”, Jesse Driscoll and Zachary C. Steinert-Threlkeld use social media data to reconstruct how the Russian-state narrative was received by Russian-speakers living in Ukraine during the critical period between February 2014 (when President Yanukovych fled Kyiv) and the Battle of Ilovaisk (when the Russian military intervened directly and froze the territorial front lines). During that time, policy elites in Moscow would have been considering using their conventional military to move the undeclared front lines of the war farther West. These planners would have been hungry for information on the social attitudes of Ukrainian Russian-speakers (russkoyazychnoe naselenie) and would have wanted to know if they were interested in opting out of the Ukrainian polity. The paper shows that Russian-language social media traffic could have been one new source of military intelligence. The paper reconstructs a number of different maps of social attitudes shared by Russian-speakers active on social media. Our dataset contains approximately 7 million tweets generated within the territorial borders of Ukraine. Aggregated patterns in the data we analyze provide a measure – noisy, but informative – of how many self-identified Russians living in Ukraine would have favored border revision. Most did not.