Twitter polling data as good as traditional methods

Here’s an interesting use of social media, one that seems obvious once it’s suggested, but not one I would have imagined. By aggregating large numbers of Twitter “tweets” researchers are able to relatively quickly measure broad approval or disapproval of ideas or public figures.

Science Magazine reports:

“A common complaint among people sucked into the ‘Twittersphere’ is that no one reads their posts. A team led by Noah Smith, a computer scientist at Carnegie Mellon University in Pittsburgh, Pennsylvania, didn’t have time to read them either. But at least the researchers put that massive pile of verbiage to good use: With an average size of 11 words, the 1 billion tweets posted in 2008 and 2009 add up to an impressive cultural data set. The team used text-analysis software to detect tweets pertaining to various issues—such as whether President Barack Obama is doing a good job—and measure the frequency of positive or negative words ranging from ‘awesome’ to ‘sucks.’

The results, which will be presented 25 May at a computer science conference in Washington, D.C., were surprisingly similar to traditional surveys. For example, the ratio of Twitter posts expressing either positive or negative sentiments about President Obama produced a ‘job approval rating’ that closely tracked the big Gallup daily poll across 2009. The president’s approval slumped over the course of the year in both. (The correlation between the two was an impressive 79% when the Twitter data was averaged across chunks of several days.)

By tracking the frequency of positive or negative sentiments about people’s financial well-being—filtering for posts about saving and spending—the Twitter data also reproduced trends in some classic economic indicators such as consumer confidence.”

Read the full article here.

Google probably has some similar sort of capability given the number of blogs, email and searches that they can data mine. I wonder if they’ve thought about monetizing the results and trying to compete with traditional polling agencies.

Or I wonder if traditional polling agencies have thought about reaching out and buying access to the Twitter Fire Hose feed (the full stream of posted data) as a way of supplementing their research given the difficulty in reaching young people by phone. (A common concern in the last presidential election.)

At the very least, a concurrent analysis of the twitter and social media feed could provide a useful “sanity check” of the traditional political poll; something that would be valuable when the results you find are surprising.

It would also be a relatively inexpensive way for organizations like the Church to be able to get access to broad based feedback to initiatives it undertakes. Say, for instance, you wanted to know if the Environment was a major concern for people in the geographic region of the diocese… or whether children’s education outweighed concerns about the divorce rate. In a time of limited resources for evangelism research, this could really be helpful.

Of course that would suppose that someone opened up a public API to the data and/or its analysis. And on whether or not some interested programers would be willing to open source some analysis tools – and maybe a simplified front end for their use.

But that’s not too far-fetched a scenario to imagine. It would be in keeping with other open-source initiatives.

Neat idea to think about. To what sorts of things would you want to put such a resource to bear upon?

Author: Nick Knisely

Episcopal bishop, dad, astronomer, erstwhile dancer...