Sure, we GeekDads can blog, but can we microblog, too? An ongoing visualization project by developer Anthony Starks reveals some interesting insights about how we use Twitter.
The project—TweetFreq—originated last May after Starks noticed certain profile pictures (ahem and ahem) showing up often in his Twitter timeline. Starks, a consultant with interests in visualization and design, worked through the Twitter API to develop a Processing application to show this behavior in a new and interesting way. Since then, Starks has generated over five dozen graphs comparing how different people use twitter. In December, Starks iterated TweetFreq again and created wtf, a web server written in the new Go programming language.
The tool is particularly helpful when looking at groups, such as popular elected officials, celebrities, influential journalists, and CEOs. TweetFreq adapted quickly to the new lists feature on Twitter, allowing Starks to take any curated list and turn it into a graphic.
The top 10 contributors to the GeekDad blog are also heavy Twitter users (source: Anthony Starks)
To put this to the test, I asked Starks to look at the recent tweeting behavior of our active GeekDad contributors. Specifically, I sent him a list of the 23 authors who posted articles in December and had a known Twitter account, rank ordered by total number of published blog posts. The top ten most prolific boasted 40 or more articles since last April. For good measure, I threw in the main GeekDad account.
“Overall, this is a nice group to show how people use twitter,” says Starks.
He points to the daily rhythm visible in the graphics (see above and below), particularly noticeable with fitzwillie (Ken Denmead) and cebsilver (Curtis Silver). “The GeekDads are fairly prolific,” notes Starks.
The results do show some visible distinction between the most active contributors and their supporting cast. The first group—which accounted for 801 of 1,130 GeekDad blog posts (70.9%) over the last 8 months—uses Twitter more frequently, averaging about 63 tweets each per week compared to just 30 for the other thirteen authors. While this sample came during the holiday season, likely curtailing some online behavior (it did mine), the data suggests an interesting insight: Tweet more, blog more.
The supporting cast for the main GeekDad bloggers don't seem to tweet as much. (source: Anthony Starks)
Using other tools in the Twitter ecosystem (When Did You Join Twitter and Twanalyst), more distinctions seem to arise. The more prolific blogger group creates tweets with slightly higher readability and conversation, retweeting more but posting fewer links. On average, the Geekdads mention other twitterers about one-third of the time and include a link 1 in every 4 tweets.
Less clear is a difference between the adoption dates for the two groups (see below). Four of the ten GeekDads who created their accounts prior to Twitter clearing 1 million accounts in Spring 2008 were Top 10 contributors to this blog. Four of the top authors signed up for the service this past spring.
Some of the power bloggers are Twitter n00bs. (image: Kevin Makice)
For more on the Starks’ TweetFreq project, check out the presentation he published in early November and an explanation of the wtf server setup.
My apologies to authors Brad Moon, Moses Milazzo, Daniel Donahoo, and Tom Angleberger, whose Twitter accounts I couldn’t confirm, and the other GeekDads who didn’t post in December, for their exclusion from this report.
