Tech

Twitter’s new Political Index tracks sentiment toward Obama, Romney

Matt Pitchford Contributor
Font Size:

Twitter launched a new tool Wednesday, the Twitter Political Index, that will track the Twitter-verse’s feelings about Obama and Romney in real time.

The Twitter Index, in partnership with real-time search engine Topsy, will sort through 400 million tweets a day from 140 million active users and use a “sentiment analysis” in order to determine which candidate the tweet is about and whether it is positive or negative.

While the computers naturally have trouble identifying irony or sarcasm, Buzzfeed reports that Topsy’s algorithms “agree with a randomly selected human 90 percent of the time on what a tweet means, validated over 30,000 tests.”

Buzzfeed also notes that the Twitter graph seems to match the ebbs and flows of Gallup polls, which suggests some degree of consistency between the metrics.

Users, however, won’t immediately be able to identify why particular spikes in popularity occur, since there is no way to find related news data inside the Index itself.

Topsy’s chief scientist Rishab Ghosh addressed concerns about whether the users on Twitter represent an accurate cross-section of America by saying “while it’s true that Twitter and Internet usage is different from general population, it’s not true to say that this is somehow unrepresentative.”

“The demographics are very very broad based. It’s a large proportion of [the] US population.”

Ghosh told USA Today, “This is not an alternative, it’s not a replacement for opinion polls. It’s a new sort of information that there was no way of accessing. We aren’t asking anybody anything. People are saying things on their own.”

It does seem, however, that this new information probably won’t be predictive of the election outcome this November.

Marc Smith, founder of the Social Media Research Foundation told USA Today that, like with photos of a political rally, “descriptions of social media discussion spaces can be important news, but we do not usually ask crowd photos to predict outcomes, even when they are newsworthy.”

Follow Matt on Twitter