Skip directly to content

Using social media to track job losses in real time

NSF Award:

NCRN-MN: Linking Surveys to the World: Administrative Data, the Web, and Beyond  (University of Michigan Ann Arbor)

Congressional Districts:
Research Areas:

A team of researchers at the University of Michigan has developed a method to use real-time data from Twitter to estimate job losses in the economy. The Social Media Job Loss Index (SMJLI) tracks losses based on Tweets and is a useful predictor of official data on claims for unemployment insurance. It can also provide additional information on the impact of specific events such as the government shutdown or Hurricane Sandy.

Economic indicators produced by the federal government rely on surveys or other records that may not keep pace with rapid changes in the economy. Although official economic indicators are a standard in terms of accuracy and coverage, they can take weeks, months or longer to collect and publish. Policymakers are left to make significant economic decisions with limited data on the current state of the economy and the labor market.

Drawing on job-related Tweets such as "lost my job," "laid off" or "canned," the index can broadly track official estimates of job loss. This research provides another measure of economic activity that may help identify turning points in the economy. It shows the potential for using data created by individuals through their use of social media to provide novel measures that, when combined with official statistics, can provide policymakers, market participants and the public with improved ability to track economic activity.

The research is part of the NSF-Census Research Network (NCRN), a partnership of NSF and the U.S. Census Bureau to support interdisciplinary research and educational activities on methodological questions of interest and significance to the broader research community and the Federal Statistical System, particularly the Census Bureau.

To monitor weekly updates of the SMJLI, visit the University of Michigan Economic Indicators from Social Media website.



  • a university of michigan job loss index predicts unemployment claims using job-related social media messages
The University of Michigan's Social Media Job Loss Index uses job-related social media messages to predict unemployment claims.

Recent Award Highlights

screen image of text interview on a mobile device

Better survey results through text messaging

Compared to voice interviews, text messaging provides higher quality data

Research Areas: People & Society Locations: Michigan New York
blood donations

Attracting potential blood donors

Economic incentives motivate blood donation

Research Areas: People & Society Locations: Michigan Ohio