Social Networks and Democracy

The following guest post is from political scientists Ora John Reuter and David Szakonyi


Foreign policy pundits have been bullish about the ability of social media to bring democratic change in authoritarian regimes. Observers have argued that social media can literally “make history” by helping topple regimes, and democracy promoters are sinking big money into a variety of trainings with this very goal.   But in countries such as China, Russia, and Iran, where users of local social networks still far outnumber users of Facebook and Twitter, authoritarian governments have used their leverage over domestic networks to contain online opposition to the regime.

The story of Russia’s most popular social networking platform, VKontakte, illustrates this point well.  In March 2013, reports (ru) surfaced about how VKontakte collaborates with Kremlin officials to gather intelligence on opposition groups that use the site.  The most damning of the reports claimed that the site shut down opposition “groups” and misdirected message traffic between opposition figures.

Indeed, in the aftermath of the December 2011 parliamentary elections, when allegations of massive electoral fraud brought tens of thousands of Russians onto the streets in the largest anti-regime protests since the fall of the Soviet Union, the relationship between VKontakte and the Kremlin even became coercive. Four days after the election, the company’s founder Pavel Durov reported that he had been summoned by the FSB (Russia’s internal security service) to answer questions about opposition activity on Vkontakte. Durov’s hesitation to cooperate fully appears to have landed him in hot water, as investors with ties to the Kremlin recently purchased a 48% share in Vkontakte and Durov may have fled the country after his home was searched in connection with an alleged traffic violation.

In our research on social media, we have found that the ownership structure of social media matters greatly for politics. When nondemocratic governments have leverage over the content and structure of social networks, users lose the ability to access independent points of view and learn about government malfeasance. Not only is information sharing monitored and potentially blocked, but democracy activists avoid networks connected with government authorities for fear of reprisals.

Though scholars have long warned about the attempts of authoritarian leaders to influence the internet, little empirical evidence has been brought forth about the effects of these efforts on politics at the micro-level. In a forthcoming article, we used survey data from the 2011 parliamentary elections in Russia to examine how usage of different social networks affected users’ awareness of electoral fraud. Our results indicate that users of Western networks like Facebook and Twitter are about five percentage points more likely to believe that there was significant electoral fraud during the elections.  Usage of Russian networks, VKontakte and Odnoklassniki, meanwhile had no effect on awareness of electoral fraud.

We argue that the reason for this discrepancy lies in the type of information being spread on these networks. During the election season, local networks’ vulnerability to state pressure seems to have led many opposition activists to focus their social media strategy on Western social networks, such as Facebook and Twitter, which are much harder to monitor and pressure.   Alexei Navalny, Russia’s most popular political blogger, maintained an active public Facebook page and Twitter account, which he used to spread hundreds of YouTube videos, photographs, and anecdotes documenting electoral fraud, and yet Navalny maintained only a token presence on Vkontakte and no presence on Odnoklassniki.  This strategy is at odds with the goal of reaching a mass audience since Odnoklassniki and VKontakte each have five times as many users as Facebook (only 5% of Russian internet users are on Facebook).

Caption: Figure 1 shows the week on week change in activity on social networking sites in the weeks surrounding the elections.  There were large spikes in activity on Facebook and Twitter, but no such spikes in VKontakte and Odnoklassniki usage.

Of course, it’s possible that individuals with preformed opinions about electoral violations select into usage of Facebook and Twitter and eschew usage of native social media platforms. Its hard to dismiss this possibility, but our findings do indicate that Facebook/Twitter users are remarkably similar to VKontakte users across a range of factors that might be correlated with perceptions of electoral fraud (sex, income, education, place of residence, support for Putin, levels of political participation, and support for the opposition).

Our presumption was also that Facebook and Twitter usage would also increase levels of protest participation, as the emerging narrative suggests.  This should certainly be true if the self-selection process described above was at work (users with preconceived notions about rampant fraud should be especially likely to join protests against electoral fraud).   But surprisingly, we found no relationship between usage of Facebook/Twitter and participation in post-election protests.

Thus, users of Western social networks were not more politically active than either their counterparts on Russian social networks or even non-users of social networks. Yet they were more informed about the wrongdoings of the government.

Our findings corroborate a slew of recent work that emphasizes an information-centric view of social media, rather than one geared towards organizing collective action. In a clever field experiment, Catie Snow Bailard (gated) randomized free access to Internet cafes in the run-up to Tanzania’s 2010 election. Again, Facebook users were more likely than the average Internet user to think the elections were conducted unfairly.  This finding aligns with recent work done by scholars on thousands of tweets emanating from the Occupy Wall Street protests in Greece, Spain and the USA. They find that online social media are used less for protest organization than they are to spread information about the grievances

What this discussion suggests is that social media’s effects on democratization are not straightforward. The information spreading function of social media is limited when most of a country’s online social networking occurs on domestic platforms that are vulnerable to government pressure, as is the case in a clutch of the world’s most prominent authoritarian regimes.  Indeed, one might speculate that one of the reasons that Russia was able to overcome its protest movement was that it was able to contain and control online dissent, while Egypt, which had no domestic social network, was unable to control the spread of information on Facebook and Twitter.  But a note of optimism is warranted:  Facebook is on the march. Just three years ago domestic social networks still dominated in much of the developing world, but the list of countries where Facebook is not dominant grows smaller every year.



3 Responses to Social Networks and Democracy

  1. Will May 18, 2013 at 4:11 am #

    FYI the unnamed ‘scholars’ who studied ‘thousands of tweets emanating from the Occupy Wall Street protests in Greece, Spain and the USA’ are:

    Yannis Theocharis, Will Lowe, Jan van Deth, and Gema García-Albacete, Using Twitter to Mobilise Protest Action: Transnational Online Mobilisation Patterns and Action Repertoires in the Occupy Wall Street, Indignados and Aganaktismenoi Movements (February 18, 2013). Available at SSRN: or

  2. Fedor Skuratov May 18, 2013 at 10:13 am #

    There is a huge mistake in conclusions — most of russian social-media recources, such as, are not controlled.
    The fact that russian Facebook users are more active in political life have much more simple explanation: core community of VK is users from 18 to 25 years old. And FB users are from 25 to 40, media persons, journalists, “creative class” (dismissive term “creacl”).

    And most of them have at least 2 accounts – one in FB (for work AND for political discussions) and one in VK (personal). It’s not because censorship or something like that, just trend (dying).

    • John Reuter May 20, 2013 at 11:05 am #

      Thanks for this comment, Fedor.

      It gives us a chance to mention something that we neglected to mention in our post. We failed to note that we controlled for a range of factors in our analyses, including age, sex, income, urban/rural status, employment status, turnout, support for Putin, support for opposition parties, and consumption of foreign media. So the results we report in the post are conditional on these covariates. But, as we admit, its an observational, not experimental, study.

      It’s also worth mentioning that the survey indicates that there is not anything approaching perfect overlap between Facebook and VKontakte usage—as of December 2011 only 10% of VKontakte users use Facebook and just over 48% of Facebook users use VKontakte. We included those who used both networks in our analysis in addition to those who used either or neither.

      There are some important differences between Facebook and Vkontakte users, but they are also similar in many respects. Facebook and Vkontakte users have similar income and education profiles and are equally likely to hail from cities, consume foreign media, and turn out in elections.

      It is true, as you note, that Facebook users are slightly older (the average Facebook user is 7 years older). They are also slightly less likely to support Putin and are less likely to support the opposition. Interestingly, they are also more likely to be unemployed (despite the fact that they are older). Again, though, its worth noting that we control for these factors in our models.