Approaches to Studying Net Freedom: Freedom House & The OpenNet Initiative

Yesterday Freedom House, which conducts comparative studies of democracy around the world, released Freedom on the Net 2011: A Global Assessment of Internet and Digital Media. The 400-page tome is composed mostly of detailed reports of 37 countries, “a representative sample with regards to geographical diversity and economic development, as well as varying levels of political and media freedom.” It also includes an overview essay and some pretty info-graphics.

It is clear that a lot of work by very bright and serious people went into this report, but I am not sure that the total is greater than the sum of its parts. The overview essay did not reveal much new information, touching on the key elements of online censorship that are already well-known: a closing of the Internet following the hayday of online freedom in the mid-nineties, offline persecution to match online blocks, implementation of censorship by ISPs. It’s a good summary, but doesn’t break any new ground.

It is inevitable that comparisons will be drawn with the OpenNet Initiative, the multi-institutional consortium that is currently the institution of record for tracking online censorship around the world. At last count, ONI tracks sixty countries, nearly twice an many as the Freedom House study. They do, however, track Internet freedom (or lack thereof) in different ways. Here are their respective indices for cross-country comparison:

The OpenNet Initiative’s Indices:

  • Political: Filtering of Web sites that express views in opposition to those of the current government, also human rights, freedom of expression, minority rights, and religious movements.
  • Social: Filtering of material related to sexuality, gambling, and illegal drugs and alcohol, other topics that may be socially sensitive or perceived as offensive.
  • Conflict/Security: Filtering of content related to armed conflicts, border disputes, separatist movements, and militant groups.
  • Internet Tools: Web sites that provide e-mail, Internet hosting, search, translation, Voice-over Internet Protocol (VoIP) telephone service, and circumvention methods.
  • Transparency of filtering policy.
  • Consistency of filtering policy over time.

Freedom House’s Indices:

  • Obstacles to Access: insufficient infrastructure, price, event-specific filtering, business access, freedom of Internet regulators.
  • Limits on Content: Extent of filtering, content deletion, transparency, user self-censorship, influence of government and interest groups on accessible online content, economic constraints, diversity of viewpoints online, mobilization use.
  • Violations of User Rights: legal protection of freedom of online speech, laws punishing online activities, detentions, restrictions on anonymous communication, surveillance, ISP cooperation with the government, extra-legal intimidation, technical violence/cyber attacks.

The OpenNet Initiative’s indices currently tracks only “filtering,” the blockage of online content. They divide the types of content blocked into three types: political, social, and conflict. They also have two separate macro measures on how filtering is administered by governments, asking if that policy is transparent and consistent.

Freedom House is more ambitious in the phenomena they seek to record in their indices, but this diversity of scope leads to somewhat muddled indices (see “Checklist of Questions” at the end of the report). One has the distinct sense that they started with a range of phenomena that they wanted to measure and then were forced to divide them into three indices, like the Freedom House’s flagship annual study, Freedom in the World, which has two indices, political rights and civil liberties, which result in a freedom status score.

There is certainly a rich range of phenomena being tracked, from self-censorship of activists to the freedom of the bodies that regulate the Internet, but the way that quite different factors are combined muddies the final score. If one is looking at an Obstacles to Access score of 15, does that reflect poor quality infrastructure, high prices for a dial-up subscription, or that the government turned off the Internet during the last election? Though ONI measures less they measure it more clearly. Their scores track less factors, making the connection between the score and the phenomena clearer.

I am glad that Freedom House is now tracking Internet freedom. Last year they tracked 15 countries and this year they tracked 37, indicating that they one day hope to create an Internet Freedom index with a global scope, perhaps similar to their Freedom in the World surveys. However, I still think they have work to do in creating clear and compelling indices to measure Internet freedom. Either more indices or fewer factors may be the way to go.


4 thoughts on “Approaches to Studying Net Freedom: Freedom House & The OpenNet Initiative

  1. Nice post, Mary. Was the first report last year or 2009? I was recently trying to find a 2010 report and couldn’t. Also, read a recent report from CIMA: Evaluating the Evaluators (, which is about the three major press freedom indices and biases. Would be interesting to compare the research and publication methodologies and see if there’s anything to be learned. I think both could rethink ways to update the country profiles more frequently and maybe leaning on people in country, where appropriate, to help. Many of ONI’s profiles, for example, are from August 2009.

  2. Nice comparison, Mary. Another point I would add is that, while I agree that ONI is well-complemented by Freedom House’s additional categories for analysis, ONI’s methods of testing filtering are empirically rigorous, employing testers in each country that we study and looking at curated lists of well over 1,000 sites in each.

    (n.b. for readers: I work for ONI).

  3. Pingback: Approaches to Studying Net Freedom: Freedom House & The OpenNet Initiative », News, Augmented

  4. Pingback: New Resources on Digital Repression | meta-activism project

Leave a Reply

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


Proudly powered by WordPress
Theme: Esquire by Matthew Buchanan.