A new study from The University of London has found that tweeting from phones makes us more egocentric.
It’s a given that Twitter personalities differ from person to person, but what about from device to device?
A new study compares the nature of what and how we tweet from our phones versus other web platforms. The research served to answer questions regarding the egocentricity, feminine/masculine language, negative/positive sentiments and agentic/communal traits of tweets and whether they occur more from mobile phones compared to other website platforms.
Led by University of London, 235 million tweets were analysed over the course of six weeks in the summer of 2013.
Here’s what they said about egocentricity:
Ultimately, we found that mobile tweets are not only more egocentric in language than any other group, but that the ratio of egocentric to nonegocentric tweets is consistently greater for mobile tweets than from nonmobile sources.
Mobile tweets are around 2.5% more egocentric than non-mobile users on average, according to the study. Mobile tweets have more egocentric language at all times of the day. Across all platforms, tweets are at least egocentric in the late mornings, around 9 a.m. to 10 a.m, and then the egocentricity levels become more moderate during work hours. The study does note, however, that overall tweet volume is known to steadily increase throughout the day.
Though tweeting is not suppressed midday, our egocentricity is. When we are at work or school, we may focus more on activities that are communal or we may not have time to tweet. After school and after work hours, the focus seems to shift back to a more egocentric presentation of self.
Their findings on the other three questions?
We did not find that mobile tweets were particularly gendered. Regardless of platform, tweets tended to employ words traditionally associated as masculine. We did find that negative language is used more frequently by mobile users at any point in time, a finding that would benefit from further research. The ratio of negative to positive unigrams was also found to be consistently greater for mobile tweets than web tweets. Lastly, we did not find that mobile-based tweets are more agentic than web-based tweets. Rather, both platforms tended to employ language that was associated with communal behaviors.
Overall, a major argument the researchers hoped to make with this study was that the context of social media content matters. While previous scholarly work has studied Twitter data, not much has looked at the differences in how they’re produced.