Who’s ahead in popular opinion in the US presidential race? Plenty of third-parties have tried to measure this in the past using tweets. Now Twitter is doing the interpretation directly through its just-launched Twitter Political Index.
Twitter blogged about the new service today, saying that it’s based on analyzing two million tweets per week and built in conjunction with the data analysis team at Topsy and two polling firms, The Mellman Group and North Star Opinion Research.
The index is updated each day (after 8pm Eastern Time) and tries to evaluate the “sentiment” expressed in tweets about each candidate.
Right now (well, for July 31), President Barack Obama has dipped four points to a score of 34 since the previous day, while Republican candidate Mitt Romney is up by two points. But with an overall score of 25, Romney still trails Obama.
The index also maintains a trend over time, so you can see how opinion of each candidate has gained or dropped. It stretches back to May 1:
Sentiment analysis can be tough. It’s easy to misinterpret what seems to be a favorable opinion that’s actually a negative one (sarcastic tweets, for example). But Twitter says that its new index compares well to traditional polls like those run by Gallup, offering up this two year comparison:
In general, the trend lines align. When they don’t, Twitter says it may be capturing “nuances” that traditional polls may miss:
The trend in Twitter Political Index scores for President Obama over the last two years often parallel his approval ratings from Gallup, frequently even hinting at where the poll numbers are headed.
But what’s more interesting are the periods when these data sets do not align, like when his daily scores following the raid that killed Osama bin Laden dropped off more quickly than his poll numbers, as the Twitter conversation returned to being more focused on economic issues.
By illustrating instances when unprompted, natural conversation deviates from responses to specific survey questions, the Twitter Political Index helps capture the nuances of public opinion.