Marco Toledo Bastos is a man with a lot of data on Twitter and how it is used as a stream for accessing news from mainstream journalism brands in a variety of countries. He is a visiting research research fellow at the Department of Media and Communications where this week he gave an presentation on his latest findings*. Here he gives us a flavour of the data which I found a fascinating way of trying to understand Twitter’s new role as a news channel.
UPDATE: You can now read the full academic journal article by Marco here
In the last four months I’ve been using the Twitter Streaming API to track the circulation of news articles of the largest national newspapers in the United States, United Kingdom, Spain, Brazil, and Germany. This is a lot of fun, and the research idea came after Twitter announced the plan to create a news service for major news organizations. It is interesting to see the sheer number of news articles that circulated on Twitter in the period per country.
I must say the data I’m posting here includes only the first two weeks of October 2012, with a total of 2,842,699 tweeted news articles. The breakdown of this figure is the following: 35% are from American news outlets, 28% from Spanish, 19% from British, 15% from Brazilian, and 4% from German newspapers.
Distribution of tweets with links to news articles from American, British, Spanish, Brazilian, and German newspapers
The charts below show the difference between the percentage of printed newspapers’ circulation and the percentage of news articles on Twitter – country by country. The leading newspapers in their respective countries also have a much higher-than-average number of news article links on Twitter, particularly El País in Spain, New York Times in the United States, Estado de S.Paulo in Brazil, Die Welt in Germany, and The Guardian in the United Kingdom.
I also looked beyond the attention share (sheer number of links to news articles) and checked the circulation of highly retweeted and mentioned messages, the relative occurrence of overlapping users and the concentration of links to news articles posted by top users.
Seemingly the proportion of messages that mentions another user (AT) varies little across news outlets of different countries (σX=0.3%), but the number of replicated messages (RT) varies considerably (σX=7%). It is significantly higher in Spanish news outlets (44%), significantly lower in German newspapers (24%), and fairly the same in the other countries (σX=1%).
Another interesting thing to take a look at is the timeline in which the news articles were circulated on Twitter. Because I was dealing with countries that span multiple time zones, I adjusted all zones to UTC+0 (GMT), using EST time zone as reference for the United States, Brasília for Brazil and the official times for the UK, Spain and Germany. The timeline of posts through the day shows a sharp growth of American links to news articles as early as 06:00 followed by subsequent peaks at 09:00, 14:30 and again at 17:30.
Links to news articles in the UK, Germany and Brazil follow a similar pattern, though the peak is not as accentuated as we see in the United States. Spanish links to news articles present a slightly different pattern, increasing steadily from 06:00 to 11:00 and decreasing abruptly after 12:0, likely due to Spanish institution of midday sleep (siesta).
There are interesting similarities and differences regarding how users tweet news articles in each countries. Social media managers and news feeds, which require minimum to no interaction from users, are widely used in Brazil and Germany, while mobile platforms like Mobile Web or Twitter for iPhone and Android are more popular in the US, UK and Spain. It’s interesting to note that these countries also have a much larger number of messages with news article links, but not necessarily a larger user base.
To conclude, I checked the top news outlets per country by number of tweets with links to news articles.
* Bear in mind that this is a research exercise that is exploring methods of researching social media such as Twitter – it is not meant to be a comprehensive survey of media outlets – so some organisations, such as Mail Newspapers in the UK, are not included.
This article is by Marco Toledo Bastos – you can contact him via marco@toledobastos.com and follow him on Twitter @herrcafe
More on Twitter as a tool for journalism at our annual journalism conference:
One quick point of clarification. The Spanish “institution” of siesta is, despite common belief, not so much an institution among the working population. Lunch break regularly starts at 13.00/14.00. Hardly you could go to get a nap at 12.00 in that case. Even though this is not related to the topic of discussion, the argument drives a wrong conclussion from the pattern and paints an old fashioned image of the real life in Spain.
“The charts below show the difference between the percentage of printed newspapers’ circulation and the percentage of news articles on Twitter”
Percentage of what?
I read this article completely about the comparison of newest and earlier technologies,
it’s remarkable article.