Twitter Allows For Rapid Public Opinion Formation But Not Evolution: Study

How do the 241 million people who feed the Twitter stream arrive at an opinion about an issue or a current event? Very quickly, it turns out.

Ideally, people would change even deeply held opinions after considering relevant facts and information. And Twitter, which affords easy access to people’s experiences and thoughts, seems like an ideal medium for a person’s opinions to evolve. But just how powerful is Twitter in shaping opinions? To answer that, a pair of Chinese researchers collected about 6 million tweets sent in the first half of 2011, aiming to analyze the massive trove of posts.

Trying to read the social media tea leaves is a lot harder than you might think – especially for individuals. A person’s posts on Twitter “do not always track the real evolution process of [a] user’s opinions because a user may not publish posts all the time, and the active users that create posts frequently only comprise a small proportion of the population,” the authors wrote in the journal Chaos on Tuesday. “Even so, we can also analyze the total posts belonging to a certain topic to investigate the dynamic trend of public opinion on Twitter.”

So, in the aggregate, it’s a bit easier to take the measure of opinions. For this study, the authors tracked the shifting opinions on three pieces of electronic consumer goods: the iPhone 4, the Blackberry and the iPad 2.

Basically, what they found was that Twitter users preferred the two Apple products – or, at least, these products had more people tweeting positive descriptions of them. In the larger sense, the pair found that the online hivemind’s opinion very rapidly reaches an opinion on an event, aided in part by endorsements from influential users and groups, and tends to dig in its heels once formed.

"Once public opinion stabilizes, it's difficult to change," lead author and Beijing Jiangtong University researcher Fei Xiong said in a statement Tuesday.

On Twitter, people are much more likely to spend time trying to sway others than to tweet about how they’ve changed their mind, the authors say.

Even though Twitter has amassed many more users since the study data was collected more than two years ago, Xiong and coauthor Yun Liu think their findings would still hold true. To get better results, it may be useful to use newer algorithms that can better parse the sentiment of people’s tweets.

The pair’s findings may indicate what could be a powerful social media tool for trying out new social media strategies in everything from businesses to elections.

“By focusing on a network application, candidates or companies can analyze the characteristics and behavior patterns of their supporters and protesters to explore whether the measures they take can influence public opinion and which opinion may succeed," Xiong said.

SOURCE: Xiong, Fei and Yun Liu. “Opinion formation on social media: An empirical approach.” Chaos 11 March 2014.