Tweetminster are attempting to use Twitter to predict the election outcome. A Japanese study last year precicted the outcome with 90% accuracy.
A group of online entrepreneurs are attempting to use Twitter to predict the election outcome. Tweetminster, a media utility that aims to make UK politics more open and social, is following the success of a Japanese study during last year’s general election which found that that in around 90 per cent of constituencies the most mentioned candidate on Twitter won the seat.
Analysis of 376 British seats since January 1st, 2009 gives Labour a slight lead of 35 per cent to 34 per cent over the Conservatives with a majority of 14 predicted. The Liberal Democrats are on 22 per cent and Others on 9 per cent. But the report is keen to outline that “assuming the same 10% margin of error of the Japanese study, the range of potential outcomes would also encompass various hung parliament scenarios with Labour short of seats”.
The report also predicts that:
• An important number of seats are being closely contested in the South West between the Conservatives and Liberal Democrats;
• The Green party may prosper in Brighton and Norwich;
• Support for the SNP is declining while Angus is a particular “race to watch”; and
• The Conservatives will perform positively in the East Midlands while Labour and the Liberal Democrats will “perform better in London than recent polls have shown”.
Alberto Nardelli, Co-founder of Tweetminster, said:
“While we have a 90% accuracy rate benchmark from a similar study in Japan, we are keen to stress that this exercise isn’t a poll, it’s an experiment into predictive modelling and we cannot speculate on the level of accuracy of these predictions at this stage.
“In fact, the whole point of the experiment is to compare mentions and word-of-mouth on Twitter with election results to determine if a correlation between the two actually exists.”
UDPATE 12.06
John Rentoul on the Independent Minds blog suggests a firm “No”
21 Responses to “Can Twitter predict the election outcome?”
Timm
This study uses the context of the German federal election to investigate whether Twitter is used as a forum for political deliberation and whether online messages on Twitter validly mirror offline political sentiment. Using LIWC text analysis software, we conducted a content analysis of over 100,000 messages containing a reference to either a political party or a politician. Our results show that Twitter is indeed used extensively for political deliberation.
We find that the mere number of messages mentioning a party reflects the election result. Moreover, joint mentions of two parties are in line with real world political ties and coalitions. An analysis of the tweets’ political sentiment demonstrates close correspondence to the parties’ and
politicians’ political positions indicating that the content of Twitter messages plausibly reflects the offline political landscape. We discuss the use of microblogging message content as a valid indicator of political sentiment and derive suggestions for further research.