How people tweet during day and night can be used to gauge unemployment levels, a new study suggests.
Data analysis of Twitter activity can actually determine how many people in a given geographic area are unemployed, researchers from Cornell University said.
"The immediacy of social media may also allow governments to better measure and understand the effect of policies, social changes, natural or man-made disasters in the economical status of cities in almost real-time," explained lead researcher Alejandro Llorente.
For the study, the team analysed social media data in Spain.
They quantified individual behavioural features from over 145 million messages spanning more than 340 different Spanish economic regions.
The results showed several clear differences in day and night tweeting behaviour and patterns that can determine high and low unemployment regions.
For example, the rate of tweeting between 9 a.m. and midday on weekdays is significantly higher in areas of high unemployment.
"Tweets in high unemployment areas are more likely to contain words such as job or unemployment, while the messages themselves are more likely to contain spelling mistakes," authors stressed.
"We attempted to quantify the extent to which deviations in patterns of behaviour that follow day and night cycles, mobility patterns and communication styles across regions relate to the geographic region's unemployment incidence," the authors concluded in a Valuewalk report.
