How social media usage does and does not predict protests

This post, published today in Monkey Cage, is based on some research, with Marco T. Bastos and Dan Mercea (recently published in the Journal of Communication, see also

A storm of protests in 2011 disputed the legitimacy of the institutional status quo across authoritarian and democratic countries alike. From the Arab Spring to the Indignados, from the London riots to the Occupy movement, a heated debate has ensued about whether people’s coming onto the streets may be aided in any significant manner by the use of social media. The debate has lingered for a decade now and is revisited with every new uprising that rises to international prominence.

This debate stems from anxieties toward social media technologies that can be used for unintended, nefarious (read socially-destructive) ends. Public discourse cautions against unverified rumors, romantic radicalization, and armchair politics with no follow-through beyond the computer screen. By contrast, the decentralized architecture of the Internet coupled with autonomous action is welcomed as enabler of civic political dissent that holds the political elite accountable in an age of widespread apathy. Either assessments hinge on the notion that communication on social network sites, such as Facebook or Twitter, scales up through social contagion much faster than before.

In a new article in the Journal of Communication, we query the type of information that scaled up on Facebook and Twitter during multiple instances of political unrest. Our first case is the Indignados demonstrations that started in Spain on May 2011 as a protest against welfare cuts, the political establishment, and the financial system. The second case is the Occupy movement that started in September 2011 reportedly inspired by the Spanish uprisings and the events of the “Arab Spring.” The movement spread to cities across the United States, and soon demonstrations happened in 951 cities in 82 countries. The third and last case is the Vinegar protests that started in Brazil on June 2013. Initially sparked by opposition to bus and underground train fare raises, the political agenda rapidly shifted against the running costs of infrastructure projects associated with sports events. An estimated quarter-million protesters took to the streets of major cities across Brazil before marches turned violent.

We analyze time-series data of Twitter, Facebook, and onsite marches reported by the media. More specifically, we examine Granger causality, which is a test to determine whether one time series (social media) is useful in forecasting another (the size of protests).

Social media usage among street demonstrators varies considerably across different national contexts. We found that communication on Twitter and Facebook predicted onsite protest activity during the Indignados. Specifically, demonstrators flocked to social media to tweet their camp-outs in city squares across Spain. More strikingly, in the course of this protest, increases in the volume of posts on Facebook and Twitter predicted a corresponding fluctuation in the number of participants.

Looking at the Occupy data, we found feedback effect in online and onsite protest activity. In other words, it was possible to predict the fortune of onsite demonstrations by turning to social media and vice versa. What is more, we saw that police action on the ground received a lot of attention on Twitter. Altogether, reflecting on the Indignados and Occupy Movements, we concluded that keeping tabs on Twitter communication (and to a lesser extent on Facebook) enables one to forecast onsite protests. In the diagram below, we show the direction of Granger-causalities across the three instances of political upheaval, with social media activity depicted in blue, camped-out, injured, and arrested protesters shown in red, and protesters attending marches colored green. Arrows indicate significant Granger-causality from one series to another.


Yet, and perhaps surprisingly, we could not establish any relationship between social media usage and actions onsite in the case of the Vinegar protests. Communication on social media is seemingly capable of amplifying onsite demonstrations, as events onsite can spark reactions online and networked communication can be used to coordinate action on location. However, the unequal Granger-causalities found in the Indignados (Spain), Occupy (United States), and Vinegar (Brazil) highlight that context is key to an informed understanding of social media usage among street demonstrators. The examination of the London riots, for instance, dismissed claims placing Twitter at the heart of efforts to recruit participants. Similarly, the Vinegar protests remind us that failing to pay attention to the wider media ecology (including broadcasting media) may foreclose an explanation for why protesters often resort to a wide panoply of tools beyond social media.

Cite this blog post
Arthur Charpentier (2015, March 18). How social media usage does and does not predict protests. Freakonometrics. Retrieved May 21, 2024, from

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.