Social Media Chatter Can Predict Unemployment, Economic Conditions
Here’s yet another nifty, unexpected application of high-powered social media analytics: social media analysis can be used to predict unemployment trends, giving an early warning of economic downturns, according to new research by SAS and UN Global Pulse, a UN think tank, drawing on social media conversations in the U.S. and Ireland from June 2009-June 2011.
The project, titled “Unemployment Through the Lens of Social Media,” set out to compare qualitative information gathered from social media with unemployment figures. The researchers examined online job-related conversations from blogs, forums and news in the U.S. and Ireland (which suffered one of the steepest economic downturns in the developed world, and which also has a high rate of social media adoption) and assigned a quantitative “mood score” based on the tone of the conversations.
Proceeding from data acquisition to dynamic correlation analysis, the quantified mood scores were then correlated to the unemployment rate, revealing leading indicators that are able to forecast rises and falls in the unemployment rate. For example, the researchers found that the volume of conversations in Ireland showing a “confused” mood correlated with an uptick in unemployment three months later. Likewise, conversations about public transportation spiked about a month before unemployment.
The SAS-Global Pulse project also measured the volume of unemployment-related content dealing with other topics like housing, transportation, entertainment, alcohol, education, and healthcare, which gave details about common economic coping mechanisms, e.g., “My beer budget will obviously be cut,” or “she doubts we can afford two cars.”
Finally, the project produced an interactive dashboard that allows users to investigate the volume of conversations around unemployment as well as the coping mechanisms that are being discussed in relation to unemployment. The dashboard also identifies time relationships between unemployment, conversation moods, coping mechanisms and various macroeconomic indicators, which allow users to see patterns and predictions.