Mumbai: If you are tweeting from a mobile device rather than your desktop, you are more likely to use egocentric language in those 140 characters, reveals a new research.
The researchers—Dhiraj Murthy (Goldsmiths, University of London), Sawyer Bowman (Bowdoin College), Alexander J. Gross (University of Maine), and Marisa McGarry (University of Maine)—who published their findings in the Journal of Communication, analysed tweets to see if presentations of self are more likely to be more egocentric, gendered or communal based when users were on a mobile device or a web-based platform.
Over the course of six weeks, they collected 235 million tweets—90% of the top sources to access Twitter were coded to denote mobile, non-mobile and mixed sources. Drawing from social-psychological methods, they then studied language use in tweets by analysing the frequency and ratios of words traditionally associated with social and behavioural characteristics.
The researchers found that mobile tweets are not only more egocentric in language than any other group, but that the ratio of egocentric to non-egocentric tweets is consistently greater for mobile tweets than from non-mobile sources. They also did not find that mobile tweets were particularly gendered. Regardless of the platform, tweets tended to employ words traditionally associated as masculine.
“Very little work has been done comparing how our social media activities vary from mobile to non-mobile. And as we increasingly use social media from mobile devices, the context in which one uses social media is a critical object of study,” said Murthy in a 1 October statement.
“Our work is transformative in this understudied field as we found that not all tweets are the same and the source of tweets does influence tweeting patterns, like how we are more likely to tweet with negative language from mobile devices than from web-based ones,” he added.
People, especially the younger population, are increasingly using mobile devices to access the Internet and social networking sites.
The number of mobile-Internet users in India is projected to double and cross the 300 million mark by 2017 from 159 million users at present, according to a 6 August report by lobby grouip Internet and Mobile Association of India (IAMAI) and consultancy firm KPMG.
Previous studies, the researchers pointed out, have linked activities performed face-to-face (e.g. eating dinner) to tweets from a particular source. And there has been research that aims to classify tweets as belonging to a particular sentiment by using word lists. This, though, is one of the first study to take a look at how mobile versus non-mobile plays a part in the language used on social media.
Tweets can already reveal a lot about human behaviour.
For instance, computer scientists from the University of Pennsylvania and elsewhere linked the online behaviour of more than 5,000 Twitter users to their income bracket. They published their results in the journal Plos One.
The team took an opposite approach to what psychologists and linguists have historically done: Rather than asking direct questions, the scientists looked at participants’ social media posts—often full of intimate details despite the lack of privacy these outlets afford.
The researchers started by looking at Twitter users’ self-described occupations. In the UK, a job code system sorts occupation into nine classes. Using that hierarchy, the researchers determined average income for each code, then sought a representative sampling from each. After manually removing ambiguous profiles—for example, listings referencing the film Coal Miner’s Daughter grouped as coal miner for profession—the team ended up with 5,191 Twitter users and more than 10 million tweets to analyse.
“It’s the largest dataset of its kind for this type of research,” said Dained Preotiuc-Pietro, a post-doctoral researcher in Penn’s Positive Psychology Center in the School of Arts and Sciences, who led the research. “The dataset enabled us to do something no one has really done before,” he added in a 29 September press statement.
From there, they created a statistical natural language processing algorithm that pulled in words that people in each code class use distinctly. Most people tend to use the same or similar words, so the algorithm’s job was to “understand” which were most predictive for each class. Humans analysed these groupings and assigned them qualitative signifiers.
Some of the results validated what is already known—for instance, that a person’s words can reveal age and gender, and that these are tied to income. But Preotiuc-Pietro said there were also some surprises. For example, those who earn more tend to express more fear and anger on Twitter. Perceived optimists have a lower mean income.
Text from those in lower-income brackets includes more swear words, whereas those in higher brackets more frequently discuss politics, corporations and the non-profit world.
Lower-income users or those of a lower socio-economic status use Twitter more as a communication means among themselves, while high-income people use it more to disseminate news, and they use it more professionally than personally, the researchers revealed.
According to market researcher Nielsen, tweets about television programmes can help in predicting the show’s success. Researchers at Boston Children’s Hospital and Merck and Co. say Twitter users who say they have trouble sleeping display a signature pattern in their messages.
And, it’s distinct enough that they can spot that pattern above the noise of Twitter, they explain in the Journal of Medical Internet Research.
Further, researchers from the Queen Mary University of London studied nearly a million tweets from over 10,000 Twitter users to reveal that liberals swear more, conservatives are more likely to talk about religion, and liberals use more individual words like me while conservatives opt more for the group-oriented us. The study was published in the journal Plos One, according to a 16 September statement.