Do you tweet less when you get more followers?
That was the subject of some new research by Andrew Stephen and Olivier Toubia from the universities of Pittsburgh and Columbia respectively. Whilst they identified that commercial users, including celebrities, tend to use Twitter as just another broadcast medium, they wanted to explore the motivations behind how non-corporate people use the site.
They picked out around 2,500 Twitter users who had between 13 and 10,000 followers. Half were placed into a control group, with the researchers recording data about each account on a daily basis over a two month period. They were especially looking for how many followers they had and how often they tweeted.
The researchers then hired undergrad students to create 100 new accounts. They were given realistic sounding names, photos, locations etc. and each account followed all 99 other accounts in addition to a sprinkling of famous Twitter users. To support the illusion that these were real people, each account also sent out some banal tweets commenting on the days weather and so on.
Over the two month period, the researchers would use the 100 new accounts to begin following users in the test group, thus gradually adding 100 new followers to each user. They then monitored these accounts to see how the increase impacted upon their post frequency.
Interestingly, there was little change in behaviour at the extremes. Users with few followers didn't show much of a change in their behaviour, with the power users likewise.
Mid-raneg users were a different story however. “Users with 13 to 26 followers did increase activity,” says Stephen, speculating that these users were encouraged by the increase in followers to post more to a suddenly larger audience.
When follower numbers were slightly higher however, the opposite occured. When users had between 62 and 245 followers, the increase by 100 prompted a decrease in post frequency. The researchers suggest that these users already had a degree of status, and the increase made them concerned that they might post something wrong and thus offend their new followership.
“As they get more followers,” Stephen says, “they want to be careful about what they post.”
The findings suggest that for many Twitter users, the motivation is more to grow their status and followership than to regularly broadcast their views.
What makes these findings particularly interesting is the implications for Twitter as a medium. It's clear that commercial users will continue to post frequently because they desire to get their content out, so if the findings are true that more regular users decrease their frequency as their follower numbers grow, it suggests that Twitter will increasingly resemble a broadcast channel rather than a social one.
That trend is tempered in the short term by the constant flow of new users into the network. If that growth slows however then it will begin to resemble a river that is bogged down with broadcasted sludge.