“power law distributions are ubiquitous” (Shirkey, 2003: 36).

Twitter, a micro-blogging platform that was founded in 2006, attracts 190 million visitors per month and the users of Twitter generate 65 million tweets a day (Schonfeld, 2010). On Twitter the contribution level is low and participation is easy. The user can post messages (also called tweets) with a maximum of 140 characters that will be displayed on their profile page and the user can join this platform for free. Therefore my assumption is that the power law of contribution and participation of ‘80/20’ is more equal on Twitter. In this article I will empirically check whether the ‘80/20 rule’ is also present in my personal Twitter network.

What kind of network are we talking about in the case of my personal Twitter network? According to the social network sites (SNS) definition of Boyd and Ellison Twitter is a SNS. The three proposed key features “allow individuals to (1) construct a public or semi‐public profile within a bounded system, (2) articulate a list of other users with whom they share a connection, and (3) view and traverse their list of connections and those made by others within the system” (Boyd and Ellison, 2007: 210). Twitter meets all these three requirements.

Twitter is furthermore a scale-free network, in principle it is able to grow limitless and scale-free networks have a very high number of nodes (Dent, 2009 : 349). But a scale-free network always follows a power law distribution. When stating that Twitter is a scale-free network, it is a matter of statistics that twitter also meet the 80/20 rule. Nielsen states that inequality exists in every online community that has ever been researched (Nielsen, 2006: 3). Further Twitter will be analysed in this article as an ego-network (Knox et al., 2006: 118; Willson, 2010: 369). I am the centre (hub) of my Twitter network, but this network only consists of the 100 people (nodes) that are part of my personal network not of the whole population/network of all Twitter users.

The data of the empirical check consists of 100 nodes (the people that I follow) including me. Posting a message/tweet counts as a contribution no matter what kind of message it contains. The result of this check are three graphs, the first graph uses the variable of the amount of tweets, the second graph uses the amount of followers and the third graph uses the variable of gender.

The first graph shows a perfect power law distribution, 20% of the users account for 84% of the tweets (see Fig. 1). The total amount of the posted tweets by the 100 nodes is 427087 thousand tweets. That means that 81% of all nodes are below the average. The node (1%) with the highest amount of tweets accounts for 20 % of the tweets.

The second graph also shows a power law distribution, striking is that the downward line is even steeper (see Fig. 2). In this context 20% of the users account for 98% of the followers. The total amount of the followers by the 100 nodes is 4709358 million followers. That means that 91% of all nodes are below the average. The node (1%) with the highest amount of followers accounts for 32 % of the followers.

The last graph shows the inequality of gender of the nodes that I follow (see Fig. 3). The majority, 57%, of the nodes is male. 26% of the nodes is female and 17% of the nodes have multiple genders (when a Twitter account is used by more people, for example by a company).

The assumption stated in the beginning that the power law of contribution and participation of ‘80/20’ is more equal on Twitter is definitely not correct. All the three graphs show that the power law distribution is highly present in Twitter. Shirkey argues “that as the number of options rise, the curve becomes more extreme” (Shirkey, 2003: 38). Possibly that explains why the curve in de second graph is steeper, because the user have the option to follow millions of other users.

When looking at my position in this Twitter network, I belong in all the graphs to the long tail or the minority. This empirical approach of visualizing a network is a formal definition of a network and a form of ‘nodocentrism’. Nodocentrism assumes that only the nodes that are visible in the network have value and need to be accounted for (Mejias, 2009: 612). According to Mejias it is also important to look at the nodes that are invisible in the network, the paranodes. Mejias state that “the point is to uncover the politics of inclusion and exclusion encoded in the network and suggest strat­egies for disidentifying from the network” (Mejias, 2009: 613). In this article the concept of the paranode is disregarded, so it only provides a formal, one-dimensional perspective.

The design of Twitter has taken the statistical power laws of contribution and participation as a starting-point to transform them into normative rules (Mejias, 2009: 611). Although the business model of Twitter is not that explicit (they do not place advertisements on their website) they can sell all the collected personal information of their users. In the terms of service Twitter state “By submitting, posting or displaying Content on or through the Services, you grant us a worldwide, non-exclusive, royalty-free license (with the right to sublicense) to use, copy, reproduce, process, adapt, modify, publish, transmit, display and distribute such Content in any and all media or distribution methods (now known or later developed)” (Twitter, 2010). This quote shows that it is almost certain that Twitter uses the personal information of their users. In that way the users of Twitter are subjected by the rules and design of the Twitter platform.


Boyd, D.M., and N.B. Ellison. “Social Network Sites: Definition, History and Scholarship.” Journal of Computer-Mediated Communication 13.1 (2008): 210-30.

Dent, Chris. “Not All Practices Are Equal: An Exploration of Discourses, Governmentality and Scale-Free Networks.” Social Semiotics 19.3 (2009): 345-61.

Knox, Hannah, Mike Savage, and Penny Harvey. “Social Networks and the Study of Relations: Networks as Method, Metaphor and Form.” Economy and Society 35.1 (2006): 113-40.

Mejias, Ulises. “The Limits of Networks as Models for Organizing the Social.” New Media & Society 12.4 (2010): 603-17.

Nielsen, Jakob. “Participation Inequality: Encouraging More Users to Contribute.” (2006).  <http://www.useit.com/alertbox/participation_inequality.html>.

Schonfeld, Erick. “Costolo: Twitter Now Has 190 Million Users Tweeting 65 Million Times a Day”. 2010. 16 December 2010. <http://techcrunch.com/2010/06/08/twitter-190-million-users/>.

Shirkey, Clay. “Power Laws, Weblogs, and Inequality ” Networks, Economics, and Culture mailing list (2003). 15 December 2010 <http://shirky.com/writings/powerlaw_weblog.html>.

Twitter. “Terms of Service”. 2010. 16 oktober 2010. <http://twitter.com/tos>.

Willson, Michele. “Technology, Networks and Communities: An Exploration of Network and Community Theory and Technosocial Forms.” Information, Communication and Society 13.5 (2010): 747-64.


-       PDF with three graphs of my Twitter network

-       Excel sheet with empirical data


  1. [...] (source: http://www.annouckwelhuis.nl/twitter-and-the-pareto-principle-2/) [...]