Social Media Posts May Predict Rises In Unemployment
Social media use has become a major feature of our culture in recent years. We share virtually everything – often more than we should – with our friends, followers, and whoever else will listen out there on the internet. Trends on Twitter and other social networking sites allow advertisers, politicians, journalists, and a host of others to know what people are thinking about or talking about in real time. Our growing engagement with social media led one research group to wonder: can posts on social media sites like Facebook actually predict certain kinds of events?
That’s the goal of a recent study by Global Pulse. They, along with software company SAS, examined job-related postings on blogs, discussion forums, and social media in order to gauge whether increases in the discussion of certain kinds of topics could predict a spike in unemployment rates.
Job-related posts were assigned a mood score to determine how those who were talking about their jobs felt about them. This data was then correlated to unemployment rates in the two countries surveyed – the US and Ireland. In Ireland the study found that an increase in negative mood scores – confusion, anxiety, depression, etc. – warned of a spike in unemployment about three months ahead of time.
They also examined references to coping mechanisms – tightening budgets, cancelled vacations, foreclosures, etc. The data showed that such references tended to follow an unemployment spike by about two months in the US.
The purpose of this study was not to develop means of predicting rises in unemployment before they happened. Instead the goal is to add depth to unemployment figures by gauging the way people feel about and cope with their situations.
Global Pulse is an initiative of the UN Secretary General’s office. It is a research and analytics organization designed to use digital information to provide real-time analytics data to the UN so that the organization can better understand and respond to current events. SAS is a software company specializing in business analytics software.
The full report is embedded below. Check it out, then let us know what you think in the comments.