Defining Social with Usage Data

February 27, 2009

Jim Zemlin’s “Facebook’s In House Sociologist Shares Stats on Users’ Social Behavior” here sheds some light on what makes Facebookers tick. As with any statistical summary, one must consider the margin of error, the sampling method, and the selectivity one brings to the presentation to data. In short, these data are not definitive, just suggestive. Nevertheless, several items bounded from the page to splash in the mine run off pond where the addled goose paddles.

First, these data substantiate what I dug out in 1999 when I poked into the behavior of engineers and scientists who used bulletin board systems. Not Facebook and Twitter grade technology, but close enough for this Web log comment. In short, in 1999 those who used the fledgling social systems did to reach a relatively small number of “friends”. What did the Facebook data suggest? The article said, “…while many people have hundreds friends on Facebook, they still only communicate with a small few. Or to quote the author of the article, “Humans may be advertising themselves more efficiently. But they still have the same small circles of intimacy as ever.” This is good news for investigatorial types because skills learned in the real world may transfer to monitoring social behaviors. Where there are clicks, there will be connections especially over time.

Second, photos are a big deal. The Facebook function that notifies a friend when a new pix is on a watched person’s page is a key driver of interaction. I am fascinated by this finding because the visual hook sets deep, lasts, and really pulls attention. I think there are some interesting ways to make use of this finding, but I am sure the trophy generation wizards are busy inventing new Facebook and Flickr services to exploit this chink in the users’ armor.

Third–and this is quite magnetizing for me–users of social networks are in “broadcast” mode. As the article said, “People who are members of online social networks are not so much ‘networking’ as they are ‘broadcasting their lives to an outer tier of acquaintances…” The idea of one way messaging–microcasting–provides some predictive worms upon which this addled goose can chew.

More Facebook data, please.

Stephen Arnold, February 27, 2009

Comments

2 Responses to “Defining Social with Usage Data”

  1. Web Analytics Wednesday Review: Sinotech Group: Web Analytics Getting Social at Chinalytics Longmarch Blog on July 1st, 2009 11:37 pm

    [...] Defining Social with Usage Data (arnoldit.com) [...]

  2. tonyzhao on June 16th, 2011 9:50 pm

    Thank you for share. I’m a Chinese.