The number of times a political candidate is mentioned on Twitter is correlated with their chances of success, according to study by scientists in the US.
The researchers, from Indiana University, analysed over 500 million Tweets posted in the run up to the Congressional elections of November 2010.
They simply counted the number of times a candidate in each of the 406 contested elections was mentioned, and found a "highly significant" correlation between the volume of Tweets and the eventual outcome of the elections.
The effect was evident even when the researchers controlled for factors such as the amount of media coverage each candidate received, the racial split of the electorate, and whether the winning candidate was an incumbent.
What is interesting about the finding is that it did not include any analysis of the sentiment of each Tweet. Tweets about how much the user hated a candidate were counted alongside messages of support.
"This is evidence for the conventional wisdom that 'all publicity is good publicity'," the scientists wrote in their report.
The fact that social media mentions were a better predictor of success than coverage by traditional media shows that TV stations such as CNN do "not seem to have a significant effect".
"This study adds to the mounting evidence that online social networks are not ephemeral, spam-infested sources of information," the scientists wrote. "Rather, social media may very well provide a valid indicator of the American electorate."
Earlier this year, US public policy group Pew Research conducted a study that drew rather different conclusions. It found that there was no correlation between the sentiment expressed on Twitter and traditional opinion polls.
For example, after the second US presidential debate last year, 77% of relevant tweets expressed a positive reaction to Barack Obama's performance. But a poll of over 600 voters after the debates found that 66% believed Mitt Romney fared better.
Pew, which conducts opinion polls, concluded that this reflects the bias of Twitter.
"The lack of consistent correspondence between Twitter reaction and public opinion is partly a reflection of the fact that those who get news on Twitter – and particularly those who tweet news – are very different demographically from the public," Pew Research concluded.
"While [Twitter] provides an interesting look into how communities of interest respond to different circumstances, it does not reliably correlate with the overall reaction of adults nationwide," it said.
For businesses, the question is whether the same dynamics apply to purchasing behaviour as they do voting.
Put together, the two reports imply that simply tallying mentions – rather than using sophisticated sentiment analysis – may better predict the success of new products and services.
