The ISIS Twitter Census Report

March 7, 2015

The ISIS Twitter census: Making sense of ISIS’s use of Twitter

For nearly a year, policymakers and the media have been fascinated with the topic of how the so-called Islamic State, or ISIS, uses Twitter.

But for all the attention devoted to the subject, some fundamental questions have remained unanswered, either left completely unaddressed, or characterized by claims that are derived from unclear sources.

With an eye toward informing the debate over how to respond to the machinations of ISIS’s social media operation, we have written the ISIS Twitter Census, a new study that examines the scope and activity of ISIS supporters on social media in unprecedented detail, including a thorough discussion of the methodology used to create the report.

The report’s findings include: 

In October through November 2014, at least 46,000 Twitter accounts were used by ISIS supporters. 

A sample of 20,000 confirmed ISIS supporters was examined to extract demographics data, with a plurality of users apparently residing in the territories controlled by the Islamic State. The second-most common location for ISIS supporters was Saudi Arabia. 

Nearly one in five ISIS supporting accounts designated English as their primary language. Almost three quarters selected Arabic, and one in 20 selected French. 

Thousands of accounts have been suspended by Twitter since October 2014, measurably degrading ISIS’s ability to project its propaganda to wider audiences. We believe the data here should permanently lay to rest the recurring objection to account suspensions based on “whack-a-mole”—the argument that ISIS can simply replace suspended accounts without suffering any negative consequences. If suspensions are carried out on a consistent basis, they do have an effect on the targeted network. 

After an initial wave of suspensions in September 2014, thousands of new ISIS accounts were created to replace or supplement those that had been lost. However, as the suspension campaign continued, significantly fewer new accounts were created in subsequent months. 

The accounts most likely to be suspended were also those most important to distributing ISIS’s message, a group of highly organized activists who post tweets at a much faster pace than normal users. 

The process of suspension does create certain new risks. Most importantly, while suspensions appear to have created obstacles to joining ISIS’s social network, they also isolate ISIS supporters online, which could increase the speed and intensity of radicalization for those who do manage to enter the network, and hinder organic social pressures that could lead to deradicalization. 

Each of these issues is discussed in considerably more detail in the full report, which is available here. We hope that the report will improve the quality of the debate over how to handle the problem of ISIS and other terrorist groups on social media by presenting a large body of preliminary data and recommendations for further study.