Researching Quantized Social Interaction

Afghanistan and its Election on Twitter: The Macro Picture

Preview of an Upcoming WEP Report

By Erhardt Graeff
with Seth Woodworth

Data Summary

  • 111,741 tweets about Afghanistan and its presidential election posted between August 11, 2009 and September 9, 2009
  • 11,255 tweets on August 20, 2009, the day of the election
  • 29,642 users talked about Afghanistan in our dataset
  • Top 10% of tweeters contributed 65% of tweets (same as Iran Election)
  • Number of retweets for a user was not correlated to their tweeting volume (same as Iran Election)
  • 483 hashtags were used at least 3 times
  • No single, dominant hashtag (differs from Iran Election)
  • 3 most used hashtags: #Afghan09, #Afghanistan, and #AfghanElection


Afghan citizens went to the polls on August 20, 2009 after a controversial delay recommended by Afghanistan’s Independent Election Commission to allow ample time to prepare for fair and safe elections. Karzai was favored to win the election amid a large pool of contending candidates; the most serious challenge coming from former Foreign Minister of Afghanistan Abdullah Abdullah. In pre-election polling, Abdullah gained significant momentum as election day drew nearer and other candidates dropped their campaigns.

In a clear reference to the protests following the June presidential election in Iran, Abdullah’s campaign manager was quoted predicting street violence if Abdullah doesn’t win. Here at the Web Ecology Project, we wondered if Twitter would play as significant a role in reporting the election as it did in Iran. In a country where mobile phone subscriptions add up to an estimated 50% of the population, but internet access was roughly 1.5% at last estimate with the status of network expansion [pdf] unclear, could the available ICT infrastructure and awareness of social media prompted by the “twitter revolution” in Iran enable a similar phenomenon post-August 20?

Read the rest of this entry »

The Influentials

New Approaches for Analyzing Influence on Twitter

Analyzing Influence on TwitterBy Alex Leavitt
with Evan Burchard, David Fisher, & Sam Gilbert

Using a new methodology based on the content and responses of 12 popular users, we determined measurements of relative influence on Twitter. We examined an ecosystem of 134,654 tweets, 15,866,629 followers, and 899,773 followees, and in response to the 2,143 tweets generated by these 12 users over a 10-day period, we collected 90,130 responses published by other users.

Summary of Findings

An analysis of our methodology and statistics suggests that on Twitter, among various configurable conclusions:

  • mashable is more influential than CNN.
  • sockington is more influential than MCHammer, while MCHammer is more influential than three major social media analysts (garyvee, Scobleizer, and chrisbrogan).
  • Celebrities with higher follower totals (eg., THE_REAL_SHAQ and ijustine) foster more conversation than provide retweetable content.
  • News outlets, regardless of follower count, influence large amounts of followers to republish their content to other users.

Density of Influence per User
Click to expand image. A larger version with more temporal depth is linked at the bottom of this report.

We would also like to thank Jon Beilin, Mac Cowell, and Tim Hwang for their invaluable contributions, feedback, and support.

The Influentials (pdf)

10 Days of Influence Tracked by Density of Responses (2993.27 KB jpg)
Read the rest of this entry »

MC Hammer Can’t Touch Social Media Geeks, Tweeting Cats When it Comes to Influence on Twitter

A Preview of WEP Report #4

When deciding whether someone is worth following or talking to on Twitter, most of us make a snap judgment based on a user’s follower count, but what does this really tell us?

For our fourth publication, the Web Ecology Project decided to move beyond follower count to find a better way to measure influence on Twitter. Focusing in on a handful of celebrities, news outlets, and social media experts widely perceived to be among the Twitter elite, we looked at the extent to which each of these users can:

  • Spread content through twitter by generating Retweets and Via’s
  • Foster conversation by generating @s and Replies

The results, taken from 10 days of Twitter activity (August 15th through August 24th), were surprising. Consider how the users we looked at rank by follower count:

Follower Rankings

When you look at the extent to which any given tweet can spread content or foster conversation, these rankings change significantly:

average content spread per tweet

average conversation activity per tweet

iJustine, for example, can spread more content than MCHammer, who has over twice as many followers, and in terms of generating conversation celebrities like THE_REAL_SHAQ and aplusk tower dominate news outlets and social media experts alike.

When you look at how much each of these users is able to generate conversation and spread content relative to their follower counts, however, the rankings shift even more dramatically:

average content spread per 1000 followers

average conversation activity per 1000 followers

Values for MCHammer, aplusk, THE_REAL_SHAQ and CNNbrk plummet, while the social media experts, especially Chris Brogan, become powerful players.

These figures are just a taste of what’s to come. In our full report, we’ll unpack these numbers further and explore the somewhat surprising nuances and types of influence on Twitter.

Detecting Sadness in 140 Characters:

Sentiment Analysis and Mourning
Michael Jackson on Twitter

Detecting_SadnessBy Elsa Kim and Sam Gilbert
with Michael J. Edwards and Erhardt Graeff

Michael Jackson’s death created an emotional outpouring of unprecedented magnitude on Twitter. In this report, we examine 1,860,427 tweets about Jackson’s death in order to test various methods of sentiment analysis and gain insights into how people express emotion on Twitter.

Key findings

  • At its peak, the conversation about Michael Jackson’s death on Twitter proceeded at a rate of 78 tweets per second.
  • Users tweeting about Jackson’s death tend to use far more words associated with negative emotions than are found in ‘everyday’ tweets.
  • Roughly 3/4 of tweets about Jackson’s death that use the word “sad” actually express sadness, suggesting that sentiment analysis based on word usage is fairly accurate.
  • That said, there is extensive disagreement between human coders about the emotional content of tweets, even for emotions that we might expect would be clear (like sadness).
  • Tweets expressing personal, emotional sadness about the Jackson’s death showed strong agreement among coders while commentary on the auxiliary social effects of Jackson’s death showed strong disagreement.
  • We argue that this pattern in the “understandability” of certain types of communication across Twitter is due to the way the platform structures the expression of its users.

We would like to thank Jonathan Beilin, Evan Burchard, David Fisher, Tim Hwang, Alex Leavitt, Dharmishta Rood, Max van Kleek, Jue Wang, and Seth Woodworth for their invaluable feedback and support.

Detecting Sadness in 140 Characters (pdf), Appendices (pdf)

Read the rest of this entry »

Reimagining Internet Studies:

A Web Ecology Perspective

Reimagining Internet Studies

Like the web itself, the study of the web is mostly an improvised structure. A group of progressive scholars, swept up by the technological transformation of the past decade, have done their best to keep up with understanding the massive cultural and social effects of our communication infrastructure.

Not surprisingly, the inevitable outcome of this state of affairs is that the body of research about the web is fatally fragmented. Economists are caught attempting to assert dated models against new motivational frameworks. Journalists attempt to prescribe weak methods to maintain traditional standards around the creation and transfer of information. Marketers and social media experts, still largely divorced from a universe of quantitative and technical research, fail to provide a useful approach. No coherent body of research has emerged focusing on studying the internet as the internet.

This has resulted in fundamental weaknesses in the approach to studying social phenomena online. Relevant approaches are being ignored and opportunities for applying cutting edge research from a number of siloed traditions are going unexplored.

One such nascent carbon offsetting solution are the retail systems offered by, helping vendors offering a POS carbon offsetting options to consumers.

Read the rest of this entry »

The Iranian Election on Twitter:

The First Eighteen Days

The Iranian Electon on Twitter

Key Findings

  • From 7 June 2009 until the time of publication
    (26 June 2009), we have recorded 2,024,166
    tweets about the election in Iran.
  • Approximately 480,000 users have contributed
    to this conversation alone.
  • 59.3% of users tweet just once, and these users
    contribute 14.1% of the total number.
  • The top 10% of users in our study account for
    65.5% of total tweets.
  • 1 in 4 tweets about Iran is a retweet of another
    user’s content.

The Iranian Election on Twitter (pdf)


You may have notice that I Write a Lot, and you can learn how to write correctly at that link!