Twitter communities and the diffusion of news


Martin Stabe recently shared a very interesting analysis about the way news is distributed on Twitter (Sharing News Articles Using 140 Characters: A Diffusion Analysis on Twitter). The researchers aimed to answer three questions:

  1. How do news sources differ in terms of the volume and the extent to which articles are shared on Twitter?
  2. What is the lifespan of a news article on Twitter as determined by the time difference between first and last tweet posted using that URL?
  3. At what rate are articles shared over time?

This sentence — about the outsize influence of the BBC — caught my eye:

The reason for this influence (aside from its popularity as a credible news source) is that the main BBC news Twitter account bbcnews is supported by 2 other BBC Twitter accounts — bbcbreaking and bbcworld.

The BBC strengthened its Twitter reach through strategic use of its organisational feeds. A few months ago, Pew completed research into Twitter networks and content diffusion. Their findings (here) revealed six main kinds of Twitter networks:

News organisations typically follow the in-hub & spoke broadcast network (bottom left in the above image). By using their support network the BBC managed to create clusters around their content, and extend the conversation.

Again according to Pew, this matters because “information sources and subjects ignite multiple conversations, each cultivating its own audience and community. These can illustrate diverse angles on a subject based on its relevance to different audiences, revealing a diversity of opinion and perspective on a social media topic.”

“Cultivating its own audience and community”

In part, this is what Rebecca Watts and Edward Maughan are doing around editorial series and special reports with the social media influencer programme. They are contacting individuals who are likely to be interested in specific content, and making sure they are aware and able to share. Becky and Ed are acting as a new source of information for a pre-existing community cluster and engaging with those involved.

Article findings (distilled):

  • Over all, financial news agencies have the lowest ratio of (1) relationships between users to (2) users, indicating a higher concentration of article tweets from individuals
  • The Guardian, FT, Mashable and Ars Technica seed most of their content through a main Twitter account. Content is then spread through retweets of the primary seeding account. Conversely, the New York Times, NPR, Washington Post, Wired and Forbes have higher percentages of tweets from users at later levels (retweets of retweets)
  • The Twitter lifespan of articles from BBC, Mashable and New York Times were among the highest of those in the study. The article lifespan from Bloomberg, Forbes and Wired were among the lowest
  • BBC, Mashable and Wired have a larger number of popular articles (those for which the life span and rate of spread is greater than the average article’s half life) in the first 48 hours as compared to the other organisations
  • The FT and Wired are the only news sources for which there is a consistent growth (i.e. influx of article tweets) during the 20- 100 hour period.